The Economic Ideas of Mao Zedong: Agricultural Transformation

The People’s Republic of China which was established in 1949 under the leadership of Mao Zedong, is today a rapidly developing nation of a billion people; on present trends it is poised to emerge as the largest economy in the world in about two decades (by the year 2015), overtaking both Japan and the USA. The counter-revolution in the USSR and its rapid economic and social collapse since 1990 under the regime of “market reforms”, lends a special interest to the sharply contrasting scenario in the world’s largest remaining socialist state. The question inevitably arises: what were the characteristics of that growth strategy which permitted rapid development during the three decades 1949- 79; and what is the relation of that strategy to the apparently very different policies the Chinese planners themselves call “socialist market economy”, which seems to be compatible with continuing rapid development -so far -in China?

This article will not go into a description of the trends after the policy changes of 1979, some acquaintance with which on the part of the reader, is taken for granted. It argues rather that a strong and broad basis for China’s rapid growth on all indicators -not only economic but also social welfare indicators, currently termed “human development” -was laid in the quarter century before 1979, as part of the Maoist strategy from the early fifties of raising the rate of capital formation sharply, and of pursuing a policy of balanced growth on the basis of the use of a range of techniques of varying capital intensity. The essential institutional prerequisite for following this strategy was comprehensive land reforms followed by the formation of co- operatives and later the communes. The main thrust of the argument of this article will be that despite all the hyperboles and encomiums, as well as the criticisms and the anathematisation of egalitarianism, the real and rational content of Mao Zedong’s contribution to the economic and social development of new China continues to be insufficiently appreciated by economists; in particular it is argued that there was an important element of innovation in the Maoist strategy of reaping the full advantages of surplus labour in rural areas, on a scale hitherto never attempted, and that this was a major component of the sharp rise in the rate of capital formation in the economy as a whole without any undue restriction of the rate of rise of mass consumption

Actual Economic Surplus and Its Contribution to Investment

In a backward economy which is launching on a modern industrial growth path, the initial problem is not that the existing economic surplus is too small, but that the specific forms in which the economic surplus is produced and appropriated, are not forms which are conducive to a high rate of investment. The main forms of surplus are land rent, usury interest and commercial profit, and only a small part is capitalist profit. In short, the economy and its agrarian sector in particular is dominated by landlords, moneylenders and traders, while the capitalists are few. The former groups may invest their surpluses “productively” from their individual point of view since they get a return by rackrenting petty tenants, squeezing debtors through high interest and so on, but their surpluses are not invested productively from the social point of view for such “investment” does not of itself add to productive capacity and to output. These types of return represent income transfers from one large group of people who lack property, to another small group of people who monopolise property in the form of land or money capital, and who are not greatly interested in improving productivity. (The landlords generally consume their surplus or use it in rent yielding or non-agricultural investment, traders by definition are not interested in increasing production, and usurers are interested in transferring peasant assets which are collateral to loans, to themselves.) The problem then is, how to transform these socially unproductive forms of surplus into socially productive forms, leading to a rise in asset formation and hence in incomes. Both in China and India, countries with a long established agrarian production, the economic surplus actually produced around 1950 in the agrarian sector could not have been less than between a quarter to a third of the net domestic product contributed by that sector, while the unproductive part was about two-thirds of the total surplus.

Historically, the answer to the problem of transforming the forms of surplus, has taken two distinct paths in the sphere of agrarian relations. The first path has been the revolutionary and, hence, socially broad-based, democratic one of abolishing the categories of rent and interest by seizing the landed property of rentiers without compensation, namely confiscating it, followed by a free and egalitarian distribution to the peasants, and writing off all outstanding loans. Given their improved status, the peasants then evolve capitalist production “from below”. The second path has been the conservative, socially elitist and narrowly based one of preserving landed property substantially, perhaps taking over a small part after paying compensation to the landlords, and distributing the land thus acquired through the market, by selling to those who can afford to pay. This method automatically limits the redistribution to a minority with money, and excludes the majority of the poorer peasants who need land most. Here, capitalism of the landed elite is promoted at the expense of a broad based peasant capitalism.

In Asia, the conservative path is exemplified by the Meiji land reform in Japan during 1869 to 1873, which abolished the feudal right to rent-cum-tax of the nobility (daimyo and the samurai) only by paying them compensation, namely, the capitalised value of their rents as cash and bonds; and then taxed the farmers heavily to finance the compensation. In Japan, the democratic path is exemplified by the land reform implemented by the US occupation regime in conjunction with the Japanese authorities in 1945, under which all land with lords in excess of 1 cho(2.45 acres) was acquired and distributed at a nominal payment to the tenants, while non-resident landlords were not allowed to keep even 1 chobut had to surrender all their land.1 The insistence of the US occupation regime in Japan at that time on the most comprehensive land reform ever in the post-war period, arose from the perception that the twin pillars of Japanese militarism rested on the zaibatsu, and the prevalence of petty tenancy as opposed to owner occupation. The revolutionary path, on the other hand, was exemplified by the confiscation and free . distribution of land in Soviet Russia after 1917, and by the land reform in China after 1949. In the latter two cases the egalitarian and free distribution of land to peasant households was thought of as the successful completion of an essentially capitalist task of doing away with feudal property, and as a transitional phase to the eventual establishment of production cooperatives or collectives, in which individual ownership of the material means of production would be replaced by cooperative and collective ownership of enlarging greatly the size of the unit taking decisions and undertaking production and investment.

An interesting study by Lippit highlights the contribution made by land reform in China directly to resources for financing industrial development.2 For the year 1952, the total of incomes obtained by the propertied classes in rural areas by way of land rent, usury interest and profit, is estimated to amount to 16.9 per cent of the value added (net income) in agriculture, to which is added the tax contributed by the owner-operators which was another 2.1 per cent of the net income, giving a total of 19 per cent of value added in agriculture as the estimate of surplus, or 9.39 billion yuan at 1952 prices. (This is almost certainly a substantial underestimate given that illegal taxes and traders’ commissions which are also a part of surplus, were not taken into account possibly owing to the absence of the relevant data.) Of this total which became available to the peasants with the radical land reform entailing the removal of the rent, interest and taxation burden, a little more than half or 4.9 billion yuan was mopped up by the new government via the new revised rates of tax and through controlled terms of trade; while the remainder or 4.5 billion yuan was retained by the peasants and raised their average income.

Thus the peasants benefited, and at the same time the new state had access to resources released by the reform: this transfer from the agrarian sector to the state, expressed as a percentage of total gross and net domestic investment in the economy in 1952, amounted to 34.7 per cent and 44.8 per cent respectively, according to Lippit, which is certainly very substantial. This reveals the direct contribution of the land reform to development finance. However, by its very nature it was a one-shot stimulus to the economy; maintaining the new higher rate of investment, and raising it further over time, would not come about automatically but had to be planned for. The increased income of peasants after the reform could, in principle, have been saved and invested; but given the abysmally low standard of living of the majority of peasants under the old system it is very likely that it would be wholly consumed, for on a per family basis it worked out only to about 55 yuan in 1952. Egalitarian land reform, while removing the incubus of a parasitic landlord and taxation system, did not thus, of itself promise a rapid rise in productive investment, therefore in the resulting output growth rate and, hence, in the ability of agriculture to provide the monetised food and raw materials needed by growing industry. In effect, poverty was now being equitably shared.

It is well known that despite its much larger geographical area than India, the cultivable area has always been less in China, and cultivation practices perforce considerably more intensive. With its longer history of agricultural surplus based society than India, the availability of cultivable land per household was only three-fifths of the Indian level in 1950, though more intensive practices by way of larger application of labour days and of manures per unit area allowed the Chinese farmers to produce higher yields and also higher output per head of population. Raising productivity further when yields are already high is a far more difficult proposition than when they are low to begin with: for there is little technical “slack” left to be taken up. Any raising of yields and of labour productivity needed large investments in extending irrigated area and reclaiming land. Each farming household individually lacked enough surplus funds for investing in fixed capital, particularly the irrigation systems and infrastructure which were a precondition for higher production; nor did individual households have much tax paying capacity, while the collection of small taxes from millions of households posed a formidable problem in terms of cost and feasibility for the government, if the old structures of exploitative intermediation were not to be revived. An enlargement of the scale of producing units, say, by forming every 100 households into a cooperative on the other hand would by itself cut the problem of tax collection to 1 per cent allow the fiscal potential of this sector to be tapped.

According to the socialist perspective, land reform in itself was a bourgeois measure taking society no further than had the French Revolution nearly two centuries earlier; land reform constituted a necessary condition for further institutional change towards cooperation. The urgency of pooling peasant efforts for the purpose of investment was all the greater given the fact that the level of environmental degradation and deforestation in China was far higher as compared to, say, in India at that time, and had been aggravated by nearly forty years of commercial exploitation, civil war and external invasion before 1950. Countering the massive problems of large-scale deforestation, soil erosion and land degradation could not be realistically done on an individual basis but required the pooled effort of many hundreds of households on local projects. The abysmally low standards of public sanitation and health care, the prevalence of epidemic diseases from snail infested canals and mosquito infested water, called for a massive collective investment effort towards cleaning up the environment and initiating a health care system to provide service to those who needed it regardless of their purchasing power.

Potential Economic Surplus Arising from Under-Employment, and Its Mobilisation

So far we have discussed the problem of mobilising the actual economic surplus produced as part of the existing level of output in the economy, for the purpose of investment. There was another source of economic surplus which could be tapped for the purpose of capital formation, however, and this was what we term the potential surplus inherent in the existence of unemployment and unemployment of labour, particularly rural labour. This is usually referred to as “surplus labour” in literature; this term has produced much confusion, with the neo-classical economists linking the existence of surplus labour to inefficiency in production and zero “marginal productivity” of labour. It should be made clear at the outset that when we refer to underemployment giving rise to a potential labour surplus which can be used for capital formation, this has nothing in common with the neo-classical concept of surplus labour. A simple numerical example will illustrate the concept of potential economic surplus arising from underemployment. In rural India as well as in rural China various estimates placed this surplus at between a quarter to one-third of the total labour-force.

Let us consider a stylised example of a small village in the post-reform period, comprising 30 farming households which are identical with respect to the size of family (5 members), the number of workers per family (2 workers), the area cultivated (0.5 ha each), the equipment possessed, and the labour days worked per annum (150 per worker and hence 300 per family) to produce the only crop -rice (two quintals per family). We assume that the most intensive known cultivation practices are being used to produce this output. For the time being we abstract from the seasonality of production and assume that labour time requirements are spread evenly over the year. The total number of workers in the village are 60 and they put in a total of 9,000 work days in the year to produce a total gross output of 600 quintals of rice (which corresponds to say, 500 quintals net output after deducting one-sixth by way of seed, animal feed and spoilage).This amount of work is essential for producing this amount of output, that is, there is no inefficiency in the system and reducing the total labour days would also reduce total output. (Neo- classical economists, on the other hand, think of a situation where the total labour days input can be reduced without reducing output, that is, there being a difference between the actual labour input and the necessary labour input, this difference being “surplus labour”. Such a situation we are assuming does not exist by saying that there is no inefficiency.)

We assume that this 500 quintals net output distributed over the village population of 250 (giving 2 qtl per head) yields enough calories for working health, but it does not provide an adequate desired income for the village population even while the workers are, by definition, underemployed since each worker can get work only for 150 days in a year of 365 days. Clearly, worker would like to work more and even for a higher income, if possible. How much more? We do not know; in the absence of this knowledge we can provisionally adopt a work norm, say of 5 days a week or 250 days a year {allowing additionally for an annual two-week holiday), which given the opportunity each farm worker might be prepared to put in. The total required labour input of 9,000 days can thus be potentially performed by only 36 workers {9,000 divided by 250) instead of the 60 workers engaged, thus releasing 24 workers {60 minus 36) from rice production, for various other kinds of capital formation projects like irrigation which are needed to raise incomes in the long run. Since food output would be maintained at the earlier level, with 36 workers remaining in rice production working harder at 250 days to maintain the total labour input of 9,000 days, each of the 24 workers released could continue to be fed at the earlier average rate of 2 quintals rice per annum while they, too, worked at the rate of 250 days on capital formation projects. Thus, as far as the maintenance cost of labour is concerned, the capital formation would be costless. In this system, the potential labour surplus in terms of days is 100 {250 minus 150) per worker, or 66.7 per cent of the days actually worked. In terms of workers too, the potential labour surplus is 24 {60 minus 36) or 66.7 per cent of the total work force.

However, in practice, it is almost impossible to actually release workers for projects as long as the individual holdings are retained; pooling of production becomes a necessary precondition for converting the potential surplus into an actual one. This is because labour surplus exists in units of labour days while workers are indivisible and the withdrawal of workers necessarily has to be in discrete units of persons; one cannot withdraw half a worker from a farm. Suppose that the 30 farms in the earlier example continue to operate separately, then releasing 24 “surplus” workers means taking away not four-fifths of a worker from each of the 30 farms, which is impossible, but taking away one worker each from 24 farms. This at once means that there will be a farm specific labour shortage; the required labour input per farm of 300 days on these 24 farms can no longer be put in by the remaining single worker without over-working and violating the work norm; labour would have to be hired, obviously from the other 6 farms from which no one has been withdrawn, but no matter how underemployed the workers on the other 6 farms might be they would not work free for another farmer and wages would have to be paid. For the workers remaining behind to work harder, as well as maintain the withdrawn worker out of farm net output, and additionally payout wages, would not be a rational outcome. Additionally, there is the question of who owns the new assets created; if the state owns it, the state cannot ask people to work in creating the assets without paying wages; if it tries to finance the wages by taxing the producers in rice-culture, they will not continue to work harder than before.

In short, the normal operation of the labour market cannot deal with this problem and “costless” capital formation becomes impossible. Essentially, what is required for costless capital formation is a system of “deferred wages”, that is, everybody agreeing to work harder without an immediate rise in income, for by its very nature an asset creating project has a certain gestation period and cannot yield immediate income. Everyone has to agree to work harder now for a higher income in the future. But, how to operate such a system within the atomistic private property where it is not clear to whom the new assets will belong? As long as there is no system for ensuring that there are incentives for people to work harder without immediate payment, they will not work harder and the potential labour surplus will remain just that, with no practical use for new capital formation. All these issues were discussed threadbare by economists in the course of the sixties and seventies in India, and the consensus which emerged was that despite the majority of empirical estimates showing a labour surplus of one-fifth to one-third of the existing labour force in different regions, mobilising this surplus for costless capital formation was virtually impossible as long as everything else was unchanged. Only the completely landless labour would freely move to project work.

In China, however, “everything else” did not remain the same: the atomistic small scale units of production, the individual small farms, in which the surplus labour time was unproductively embedded, were done away with and the production pooled in larger units. It is precisely in permitting the pooling of surplus labour time and, hence, in enabling the effective mobilisation of the potential labour surplus in the form of discrete units of workers for capital formation, that the strength of the cooperatives and the communes lay. The pooling of production in the advanced cooperatives {comprising around 200 to 300 households) and later the much larger communes {about 3,500 households) did away with the practical problems of the withdrawal of workers for project work in discrete units, for as soon as production was pooled a smaller number of people working more days per year could not only put in the same total labour time in crop production as before, but indeed raised output further if the initial underemployment was severe, releasing others for projects of material and human capital formation.

At the same time, the egalitarian distribution system ensured that the withdrawn workers had access to the same basic standard of life as others {even though the project work they were doing actually yielded nothing to begin with), by giving them the same right to draw grain and necessities rations as the workers in crop and sideline production actually producing these necessities; and by allotting them work points for project work, against which they could claim a share in the remaining collective income. A policy directive ensured that up to 70 per cent of the grain output was to be distributed equally regardless of work points earned. In such a system the local availability of grain itself was crucial and a logical corollary was that all regions should grow grain.

In fact, even though the well known theory of socialist distribution was ‘rom each according to his ability, to each according to his work”, in practice the Maoist egalitarian strategy involved a deviation or went beyond this principle; what was being actually implemented was ‘rom each according to his ability, to each according to his basic need’s. This was by no means a utopian or unrealistic strategy; on the contrary, it was the only practical system which could have permitted the large-scale mobilisation of surplus labour which took place, and which served to raise the rate of capital formation to levels which have not been accurately estimated by economists to date. “Linking remuneration directly to work on the basis of household functioning would have meant that project workers would have got nothing and indeed could not have been any thing short of a disastrous policy at this stage, for what would have been induced into project work at all, except under similar systems of wage and other oullays by the state financed by inflationary credit and deficit financing, as were being operated in mixed economies like India at the time. Incidentally, all existing estimates of capital formation in China during 1955 to 1978 are underestimates because they ignore the non-monetised investment of labour which we are discussing; even those few estimates which try to take into account project labour, undervalue it because agricultural productivity is inadequately estimated.

A reading of the writings of Mao Zedong during the crucial period of the transition to the advanced cooperatives reveals the awareness of the labour mobilising potential of the cooperatives.

In 1954, commenting on a piece written by another author, Mao said, “under present conditions of production there is already a surplus of roughly one-third of labour power. What required three people in the past can be done by two after ccoperative transformation, an indication of the superiority of socialism. Where can an outlet be found for this surplus labour power of one-third or more? For the most part, still in the country-side. …The masses have unlimited creative power. They can organize themselves to take on all spheres and branches of work where they can give full play to their energy, tackle production more intensively and extensively, and initiate more and more undertaking for their own well being.”3

In the event, the masses were not entirely left to decide on the undertakings to be locally initiated, but rather local adaptations and variations were to take place within a broad overall policy of promoting undertakings in the following groups : physical capital formation via land reclamation, hill terracing, afforestation and irrigation; infrastructure (roads, bridges and buildings); energy; rural sidelines and industries; and human capital formation (public sanitation, clinics and schools). The possibilities were certainly nearly limitless at this time given the existing abysmally low levels of material, educational and health development in the countryside. On land reclamation and afforestation Mao Zedong stressed the need for “state organised land reclamation by settlers the plan being to bring 400 to 500 m. mu of wasteland under cultivation in the course of three five year plans”. He went on to say:

“I think the barren mountains in the north in particular should be afforested, and they undoubtedly can be. Do you comrades from the north have courage enough for this? Many places in the south need afforestation too. It will be fine if in a number of years we can see various places in the south and north clothed with greenery.”4

Mao Zedong confidently expected, even within the advanced cooperatives and before collectivisation, that the annual labour days employed per worker would rise substantially and that female participation rates would also rise as more rural undertakings were established, existing labour surplus thereby mobilised and increasing supply elicited:

“Before the cooperative transformation of agriculture, surplus labour-power was a problem in many parts of the country. Since then many cooperatives have felt the pinch of a labour shortage and need to mobilise the masses of the women, who did not work in the fields before, to take their place on the labour front…. For many places the labour shortage becomes evident as production grows in scale, the number of undertakings increases, the efforts to remake nature become more extensive and intensive, and the work is done more thoroughly.”5

Further, he goes on to say:

“Things in this country also show us that an outlet can be found in the villages for rural labour power. As management improves and the scope of production expands, every able-bodied man and womarl can put in more work-days in the year. Instead of over one hundred workdays for a man and a few score for a woman as described in this article the former can put in well over two hundred workdays and the latter well over one hundred or more.”6

In fact, according to P. Schran’s data from official sources, the annual number of days employed per person in rural China rose from 119 to 189 during 1950 to 1959, with most of the increase seen after 1955 during the shift to the advanced cooperatives.7 (see Table below) The importance of the virtually costless capital formation entailed in mobilising surplus labour, can hardly be exaggerated.

Peasant Labour Day Inputs In Chinese Agriculture

Note: The sharp increases from the large-scale mobilisations for rural water conservancy in the winter of 1955.56 and 1957- 58 are reflected above. The estimates cited for 1959 appear somewhat improbable as demobilisation was seen throughout the year. Source: Peter Schran, The Development of Chinese Agriculture 1950-1959, p.75.

More than anything else it served to raise sharply the rate of investment in the economy; the official estimates almost certainly understate the real extent of rise since the unmonetised part of investment, which we have been discussing, has not been included. In rural areas, capital formation could be measured, and was often expressed, not in value terms but in terms of physical indicators like the volume of earth shined in reclamation and terracing work, the number of irrigation reservoirs constructed, the miles of canals dredged, etc. Further, the rate of investment was raised sharply without imposing cuts in consumption on particular segments of the population, owing to the egalitarian distribution system. The rate of rise in real consumption of both peasants and workers during 1958-59 to 1978-79 was certainly slow, as might be expected since the trade- off in this early stage of industrialisation was between a higher rate of investment at the expense of a lower rate of consumption rise, in order to secure a higher rate of future consumption than would be the case with the alternative scenario of a higher rate of consumption now. But the egalitarian strategy essentially distributed the burden equitably.

The timing of the shift to the large-scale communes was not a happy one; it coincided with a run of very poor harvests complicated by floods in some parts of the country and attacks of pests in others. There was a very substantial downward deviation of output from the trend during 1959 to 1963, and this has complicated the evaluation of the shift ever since. There is a severe problem of causal identification here: it is arguable that even without the institutional change to communes, output would have fallen anyway, for agricultural output is subject to cyclical patterns of movement; India saw a severe downturn a few years later, in 1964-65, with no institutional change. Many scholars who are disposed to criticise the idea of large-scale collective production have, however, tended to place the main burden of the output decline on the shift to the communes; this in our view is not a tenable position. What is probably true is first, that the decline which would have taken place anyway, was exacerbated by the initial severe management problems entailed in the shift; and second, too much grain was procured during 1959-61 by the government which was unrealistic in expecting too immediate an impact on productivity of the shift to large-scale production. Given that output had actually fallen rather than increased, this led to a severe decline of rural availability of food grains, and an avoidable rise in crude death rate in rural areas.8

The egalitarian distribution system, paradoxically, must have served to mask the real impact of the food availability decline of 1959-61 since, unlike the case in a class society, this decline was no longer concentrated on particular narrow segments of the population, and therefore did not take the dramatic form of able-bodied poor people of working age showing extreme emaciation and death by disease and starvation. It was spread out much more evenly over the population of a given affected area than would have been the case even fifteen years earlier, and probably the decline in nutrition would have affected vulnerable groups like the very old, the very young, and parturient women the most. We do not believe that there was a sinister conspiracy of silence regarding the “famine”; rather, there was a genuine problem of appreciating the magnitude of the impact of availability decline in an egalitarian society. Along with a rise in the overall crude death rate, which emerged statistically much later as the numbers were put together, there also appears to have been a dramatic decline in the birth rate during 1958-60. This is perhaps not difficult to understand, for labour power was being mobilised almost on a war-for-production footing, and women were being extensively drawn into the work force at this time. In this initial period of the “Great Leap”, household activities like cooking and child minding were also briefly socialised with the establishment of communal kitchens and creches to free women for labour, and mobility increased greatly as women moved to project work. With such a reversal of the old patterns of life and work amounting virtually to social dislocation, it would not be surprising if the decision to have children was postponed, reflected in a fall in the birth rate. The decline was sharp enough for the rate of natural increase to become negative in the year 1960. From 1962 the birth rate again jumped to unprecedentedly high levels as though the postponed decisions were now being taken, resulting in a marked “bunching” of births.

Some scholars have used a very dubious method of arriving at grossly unrealistic and inflated “famine deaths” during this period (1959-61) by taking account not only of the higher crude death rate (which is a legitimate measure) but also counting the “missing millions” as a result of the lower birth rate, as part of the toll. There is a great deal of difference between people who are already there, dying prematurely due to a sharp decline in nutritional status, and people not being born at all. The former can enter the statistics of “famine deaths” according to any sensible definition of famine, but people who are not born at all are obviously in no position to die whether prematurely or otherwise. The exaggerated figures of 30 million or more “famine deaths” in China are arrived at after including the missing millions because they were not born and, indeed, were not conceived at all. On this logic we would have to talk about large scale “famine deaths” in Western Europe during two World Wars owing to the fall in the birth rate. Instead of consistent application these academically dubious concepts are reserved, apparently, for exclusive application to the developing economies. More responsible estimates place excess mortality in China between 10 to 13 million during the 1959-61 period if the increased crude death rate during these years is compared to the level of 1958. The excess toll of this order is bad and is a permanent blot on the otherwise impressive record of welfare gains in the Maoist era. The highest level of the crude death rate in China’s famine in 1960 which was 25.43 per 1,0009 was incidentally, lower than the “normal” average crude death rate during 1955-60 in eighteen developing countries; while the “normal” Indian crude death rate was very close at 24.6 per 1,000 during the same period, 1955-60. Needless to say, no one talks of “famine” in these developing countries or even in India -a good example of academic inconsistency.

After weathering the crisis of the 1959-61 period the commune system settled down, management problems were gradually overcome and the full benefits of large scale operation combined with continuing mobilisation of labour for material and human capital formation, were realised up to 1978. The crude death rate resumed its steep decline with the public sanitation campaigns and the provision of an elementary but effective rural health care system, which dramatically lowered infant and child mortality levels, while illiteracy declined more steeply than in any other developing country. Raw ski has shown that the agricultural productive base was transformed, with the use of modern industrial inputs (fertilisers, fuels, cement) and fixed capital use (irrigation and drainage equipment, tractors, power tillers and other farm machinery) growing at over 20 per cent per annum during the fifteen to twenty year period before 1978. The winter works programmes provided employment for 2 to 2.5 months in the year during the seventies to workers numbering 50 million in the early seventies, rising to 100 million during the three years preceding 1979; this accounted for 30 per cent of the labour force. The winter works covered terracing, land reclamation, earthworks, irrigation works and the energy sector. By the late seventies, the average annual employment per worker in rural areas increased up to 250 days, compared to around 190 days in 1959.10 This was a very creditable achievement given the fact that the work force was larger not only owing to growth but also to a large rise in the female participation rate.

After the policy regime shift in 1979, the employment generation and collective capital maintenance and new capital formation programmes received a severe setback as decollectivisation proceeded; the re-emergence of a serious unemployment problem in rural areas has been one consequence of the reversal of the Maoist development strategy.

Ultimately the rate of development of a society, not only economic growth as conventionally defined but also improvement in human welfare, depends on the rate at which it can invest in material as well as human capital formation. This essay has shown that the fundamental innovation of the Maoist development strategy, of which egalitarianism was an essential component, lay in converting an apparent liability into an asset: by directly transforming under-employed surplus labour into capital at minimal extra cost, a firm basis was laid for agricultural productive transformation which fed into industrial growth, as well as for the gains in human development indicators.