It is well known that the composition of India’s domestically produced output – the gross domestic product (GDP) divided into primary, secondary, and services sectors – has changed faster than the occupational structure of the workforce in those sectors. According to standard wisdom in development economics, these two changes should go more or less hand in hand. However, the two processes of the transformation in output and labour force composition analysed by Simon Kuznets and Arthur Lewis respectively, need to accommodate the possibility of significant differences in the speeds of adjustment. For example, the change in the composition of output might be significantly faster than in the occupational structure, as is indeed happening in India.
The problem with viewing this as merely difference in speeds of adjustment is twofold: political and economic, both of which now haunt Indian development. Politically, unless industry creates jobs at a sufficiently high speed, growing unemployment will tend to undermine the industrialisation process. The problem may be mitigated to an extent by higher growth of the service sector, a sort of post-industrialisation scenario of output and employment growth in a largely pre-industrial society. To what extent this is a viable model of economic development is a matter of debate.
Problem of jobless growth
The political problem of grossly inadequate industrial job creation surfaces as an unmanageably vast informal economy that exists side by side with the formal organised economy. It shows no clear sign of deceasing in importance with industrialisation, and overlaps with the same threefold division of the formal sector into agriculture, industry, and services. Not much statistically reliable information is available about the informal sector, and we must rely to a large extent on the ‘residual method’ of what is not in the formal as informal.
However, when all the rest is lumped together as the informal sector, it becomes difficult to characterise it either in terms of the nature of output or conditions of work. It is often not clear, for example, what an informal industry is, except defining it arbitrarily by the scale of operation, the conditions of work like the extent of insurance cover for workers, fiscal status of the company, etc. It is even more problematic when workers are partly formally and partly informally employed, or a worker has several employers, or works intermittently from home at piecemeal rate, and similar various other types of part-time informal employment. A particularly important form of organisation of work is when the whole family works almost as a single labour unit on a small farm or shop which usually goes under the name of ‘self-employed’. Naturally with several arbitrary definitions to classify data, different and often contradictory estimates of labour, and its pattern of participation and employment exist for the informal sector.
The political problem of grossly inadequate industrial job creation surfaces as an unmanageably vast informal economy that exists side by side with the formal organised economy.
Nevertheless, a broad picture can be formed using labour productivity as the criterion. About 90% of India’s labour force is engaged in the informal economy, producing about 46% of output. This means that labour productivity in the informal sector is slightly above half of the national average. Self-employed persons mostly in small and marginal agriculture and allied activities form about 60% of the workers in the informal sector, with about one-thirds of national labour productivity. Using weighted averages, this means the remaining 40% of the labour force has about three-fourths of the national average labour productivity.
The remaining 55% of GDP is produced by only about 10% of the labour force in the formal sector – 5.5 times the national average. And only about 5% of that labour force is in large business, mostly in various public and private corporate industries and services including public utilities.
I present these broad ‘orders of magnitude’ combining statistics from several sources about the composition of output and employment by sectors. Although not exact, they provide an overall picture, particularly of relative labour productivities, which will become useful for my argument as follows.
Note that these figures entail enormous differences in labour productivity between the informal and formal sectors, particularly between small agriculture and large corporate business. From the above figures, we may infer that the labour productivity difference between the two is about 12 to 16 times. This enormous difference in labour productivity is a double-edged sword. It is a potential source of a large increase in output by shifting labour from small agriculture to large, organised industry (as Lewis visualised), but also has the destructive potential source of creating a serious unemployment problem, because any given level of output can be produced with much less labour. Thus, it can result in a serious loss of livelihood if the much higher labour productivity in the organised sector faces the constraint of market size to which neither Kuznets nor Lewis paid the attention it deserves. The constraint of market size is caused by insufficient effective demand. Organised industry will then find it unprofitable to expand production beyond what the market will take, and fail to accommodate most of the displaced workers from agriculture.
An example can be given to drive home the point. Suppose 100 workers producing 300 units of output (i.e., labour productivity is 3) are displaced from small-scale agriculture by organised industry which has, say, 10 times higher labour productivity at 3x10=30. If the market size does not change and the level of effective demand remains the same, then the same pre-displacement level of output of 300 can be produced by (300/30), i.e., only 10 workers. This means 10 workers employed in organised industry produce the output that can be sold in the market, while 90 become redundant and unemployed. This also means that the level of effective demand has to increase by the same factor as labour productivity difference, i.e., by 10 times, to absorb all displaced labour.
When this massive demand expansion does not happen, and only a few of the displaced workers are absorbed, the rest becomes surplus labour. Here we have the first departure from the conventional wisdom of development economics taught by the Lewis model. Industrialisation typically creates and destroys employment and livelihood simultaneously. Note, also, that if some more can be employed because the market size is somewhat larger, say 330 instead of 300, to provide employment to (330/30) =11 instead of 10 workers in organised industry, there would be 10 % growth in output and employment in the industrial sector, but still with unemployment of (100-11), i.e. 89 are left unemployed from the small-scale agriculture sector. There would be growth in GDP along with growing unemployment, a typical scenario of jobless growth.
Creation and destruction of employment during the process of creative destruction are assumed to be balanced in a long run equilibrium by ignoring the problem of effective demand.
Joseph Schumpeter had visualised technical progress as the propeller of capitalistic growth raising labour productivity over time. He recognised that ‘frictional’ unemployment can arise in this process of ‘creative destruction’ caused by waves of technical progress. However, he and most of his followers did not see that this can be a persistent problem of inadequate effective demand and market size. In most typical discussion that followed, creation and destruction of employment during the process of creative destruction are assumed to be balanced in a long run equilibrium by ignoring the problem of effective demand. It is assumed away as non-existent in the long run when presumably the price mechanism has fully worked to clear all markets including the labour market. Thus, frictional unemployment becomes a transient lapse from long-run full employment growth equilibrium. However, this construction is no more than a convenient article of faith in so far as we always live in the short run. Our economic analysis particularly of the unemployment problem must be relevant to this ever-present short run.
When the problem of insufficient effective demand is recognised, it may be argued that the foreign market, particularly under globalisation, will provide sufficient demand. Foreign trade may relax the constraint of effective demand through the operation of the foreign trade multiplier, provided there is trade surplus, i.e. exports exceed imports. This argument about globalisation as relaxing the demand constraint has turned out to be systematically inapplicable in the Indian case, because India has a persistent and significant trade deficit (the deficit is reduced when GNP including net factor payments is considered, but the current account deficit is still around 1.5 % of GNP on average). As a result, the foreign trade multiplier works in reverse to subtract rather than add to the size of the Indian domestic market. Attempts to enhance international competitiveness and reduce the trade deficit through greater mechanisation and robotisation has not reduced India’s trade deficit. With the high import intensity of these processes of mechanisation, the additional import often exceeds additional export at the margin. With a persistent trade deficit, India’s international competitiveness has not increased; rather it has encouraged labour-saving production processes. This has only worsened the unemployment problem.
A special mention must be made here of the argument of labour market flexibility, mostly consisting of lower wages, longer hours of work and the enactment of laws against workers’ rights. These restraints on wages have reduced, rather than increased effective demand by depressing workers’ consumption, which has not been compensated either by higher investment or consumption of the other classes. However, the show has been kept going with international capital inflows, mostly in the nature of ‘portfolio investment’, which do not expand productive capacity of the economy, but affect the ownership pattern of stocks.
With a persistent trade deficit, India’s international competitiveness has not increased; rather it has encouraged labour-saving production processes. This has only worsened the unemployment problem.
Some top 10 to 20 percent of the Indians have been happier with a more luxurious consumption basket made possible with partially or totally imported goods, sustainable only with international capital inflows. It has not only eased the foreign exchange constraint on their consumption, but also helped in boosting prices in the stock market. This happier and politically most vocal upper class of Indians wants this show to go on. So do especially two or three large corporate houses not only because it helps them to use their inflated stock prices as collateral for borrowing from banks, apart from other undue favours they receive from the government. The unemployment situation worsens relentlessly, while the government and the corporations talk of higher GDP growth, but not of employment!
Implications of jobless growth
It is difficult to resist the temptation of digressing a little at this point into academic welfare economics, and consider the implications of this jobless growth by imagining a crude compensation test of the Kaldor-Scitovsky-Samuelson variety. In the case of jobless growth, output and employment are gained in the organised industries but output, livelihood and employment are lost typically in the unorganised sector, especially in small agriculture due to land acquisition. The gainers as newly employed in industry can overcompensate the losers (Scitovsky) only in terms of output due to their higher labour productivity when growth is positive but not in terms of providing alternative livelihood to those without jobs. The losers, on the other hand, cannot overcompensate the gainers (Kaldor) either in terms of output or employment. The Samuelson double compensation test is a sufficient condition where both the gainers and the losers can compensate one another. This sufficiency criterion fails. From this point of view, industrialisation may be justified in terms of output growth but not in terms of livelihood or welfare considerations. Therefore, unless some mechanism of compensation to the losers is put in place by the State, the path to industrialisation is not inclusive, and might face growing resistance from the increasing number of losers in a democracy.
In the context of the Indian economy, this particularly means all those displaced from small-scale agriculture without an alternative livelihood are unambiguous losers, and the human situation only gets worse in the economy unless compensation is actually paid by the State through some mechanism because higher growth fails to do that job. Indeed, the mechanism for compensation payment needs to be specified as an integral part of the industrialisation process. Without an institutionalised welfare state where basic needs like nutrition and health, education, housing, and old age pensions are guaranteed by the state, make-shift mechanisms emerge. Various mechanisms for compensating those adversely affected emerge especially at election times by promising various freebies. It gains momentum in a competitive democracy with the political parties trying to outdo each other. This invariably is a zero-sum game if total welfare funds are fixed, and the ruling party has a decisive edge. If the welfare fund has to be expanded, this requires taxing the gainers of this jobless industrialisation either by heavily taxing the rich corporates and the upper classes, or seeking temporary solutions like selling off public assets, including not only public sector enterprises but also natural resources. In India, we now have the spectacle instead of offering tax breaks, subsidies and handing over of natural resources like land, water bodies, village commons, the sea shore and forests to corporations in the name of incentivising them for higher growth.
Silence is the golden rule which no political party wants to break because none have escaped the slogan of higher corporate-led growth. As a result, no political party has been able to confront the real challenge of creating extensively decent livelihoods and employment. Pretending to bypass the problem by promising various dole-outs is a hollow electoral game played with much show and little substance so long as it lasts because the real problems of unemployment, lack of livelihood and destitution grow like a galloping cancer to engulf the system.
Problem of corporate-led industrialisation
The real issue is political. Neither our academia nor our political establishment is unwilling to face it honestly. When the job of industrialisation is entrusted to large private corporations as an essential aspect of growth, the State is hardly in a position to devise a fiscal mechanism of fair compensation to be borne by private industry. Instead, a dangerous anti-people mutualism between the government and the large corporations develops assuming the name of economic development. Symptoms of the disease are abundant: heavily subsidised price for land and related natural resources for corporations, tax breaks and cultivated blindness to frauds by favoured corporations, and receiving in return generous contributions to the party and its election funds (think of electoral bonds, wilful defaults of bank loans).
When the job of industrialisation is entrusted to large private corporations as an essential aspect of growth, the State is hardly in a position to devise a fiscal mechanism of fair compensation to be borne by private industry.
The cost of industrialisation is borne to a disproportionately large extent not by the general public but by the most vulnerable sections. High prices, rising cost of living, and heavy indirect taxes are general symptoms, but much worse is hidden from the public eye. In India, the Adivasis (the Scheduled Tribes), the poorest amongst us, and roughly 8% of the population account for 40% of the displaced population, i.e. they face five times higher probability of being displaced in the name of development. A similar story would be told if we could have reliable data of the displaced classified by caste, but in its absence we can only note that the incidence of land acquisition falls disproportionately on the small, not large holders. Around half of the Dalit households had holdings less than 0.4 hectares. Their average holding is 0.52 hectares, which is about half the average holding of other households (1.05 hectares), and only 6% of them had more than 2 hectares of land. Acquisition of land on a disproportionately large scale from them in the name of economic development deepens economic inequality and exacerbates social divisions in the countryside.
Land acquisition in the name of corporate industrialisation, and the handing over of land at highly subsidised price to the corporate houses is the most telling symptom of this toxic economic development we are undergoing, but numerically it is not the most important factor in India’s unemployment problem.
Any workable estimate of unemployment is notoriously difficult in a country like India mainly because of its large agrarian and informal sector. Looking through several estimates based on alternative definitions, and a reasonable guess on the conservative side would be that about 11 million persons currently enter the labour force every year with at least 20 million as the carry-over unemployment from the past. Assuming an average 7.5% unemployment, these figures yield an annual 4.1% growth in the population of unemployed, in addition to the growth rate of population at approximately 1.9%. This means employment has to grow at least at 6% per cent per annum (without considering labour productivity growth) so as not to let the unemployment situation get worse over time. Two qualifications are essential here. First, the really poor in search of their next meal, so to say, cannot afford to be unemployed, and in this sense the unemployment estimates used here are likely to be underestimates. This comes out clearly in all statistics which show that the more educated (correlated with higher income and caste) have a higher rate of unemployment. This is in stark contrast to developed countries ,where the relationship between the education level and unemployment rate is negative. In our case it is positive.
Theories of unemployment which are based on the idea of time spent on search and information gathering as the cause of unemployment are misleading in this sense, because the time individuals can afford to spend on search depends on their economic situation. When most people are very poor, and cannot afford to wait searching for a better job, the main explanation must be sought elsewhere. Second, a similar phenomenon is also visible in the Indian statistics in the case of women’s participation in so far as very poor women have a higher rate of participation than better off ones. This understanding is important for devising employment guarantee schemes. At near subsistence wage with relatively long hours of work, it is reasonable to suggest that those who would seek employment cannot afford to search around for long. Therefore, making such jobs available without complicated bureaucratic procedures is sensible, and satisfies the criterion of ‘procedural rationality’.
All statistics …show that the more educated (correlated with higher income and caste) have a higher rate of unemployment. This is in stark contrast to developed countries where the relationship between education level and unemployment rate is negative. In our case it is positive.
Corporate industrialisation is not the solution but part of the problem. Because large scale land acquisition for industrialisation adds to this growing pressure of unemployment with more people losing livelihoods than gaining jobs, and offers no solution in sight. So long as we continue on this path of development, the unemployment problem will only worsen. In a functioning democracy with electoral accountability, the desperate employment situation on the ground will overwhelm the mystical idea of reaching a long-run equilibrium with full employment in industry. A solution along a different pathway must be attempted for reversing this dismal prospect. This would require making conventional wisdom of development economics stand on its head.
Strategy for development
I sketch out an overall strategy, listing briefly the inter-connected central ideas as the framework for specific policies, as follows.
We do not try to increase overall labour productivity in the economy by trying to transfer labour from low productivity small agriculture to high productivity organised industry, similar to the so-called Lewis process. Instead, our focus should be on reducing the gap between the two by raising the productivity of small agriculture in the present phase of India’s development.
The emphasis would be on raising the productivity of land, and not of labour. Infrastructure development including road communication and connectivity should have this land productivity augmentation as focus, instead of the currently pursued self-defeating programmes of creating world class connectivity amongst cities amidst a sea of destitution and poverty in the countryside.
Land productivity will need to be defined in an inclusive way. All types of agricultural produce including subsistence and commercial produce, poultry, fishing and animal husbandry should define the productivity of land in an area. Forests, rivers, water bodies, medicinal plants and marine products of the commons should be treated as part of an increase in public rather than private wealth, raising land productivity. The external economies generated in the process will be partly internalised by private firms, but they will also contribute to fighting degradation of the environment and climate change.
Democracy means allowing dissent, and inter- as well as intra-class conflict of interests amongst peasants is almost inevitable. The extent to which it can be resolved at the panchayat level will be a crucial criterion of judging success…
Crop composition and inter-cropping varying with soil and climatic conditions should be a priority area of state support for traditional knowledge and field-based research. The relative price of crops is an important instrument in the hands of the policymakers. In particular, prices and other fiscal incentives should be used for intelligent water management for influencing the cropping pattern in view of differences in soil and climatic conditions. Reaching such an inclusive optimal cropping pattern has to be attempted by decreasing rather than increasing the dependence on inputs purchased from markets. This would tend to encourage the idea of local resource-based organic farming with more efficient use of water and power and access to market network developed for this purpose.
Less of a dependence on market-based inputs will have a greater impact on the cost of cultivation and the recurring problem of debt, particularly of small peasants. This must be carefully monitored by local banks to ascertain that some reduction in their agricultural debt is actually being achieved, and a bonus system can be instituted to encourage further this system. The credit policy, local warehouse facilities and marketing networks have to be synchronised with special attention given to small peasants.
Our focus should be on reducing the gap in productivity between small agriculture and organised industry by raising the productivity of small agriculture in the present phase of India’s development.
A minimum support price system for agricultural produce is an imperative of our time, both as income support and as a price incentive. A minimal level of crop insurance has to be introduced, while ensuring that crop failure, if it happens, has been in spite of the crops being grown fitting local climatic conditions and and water management requirement. An insurance premium will be collected by the local panchayat in normal years. It is hoped that under these measures of collective water, land and forest management aided by the panchayat system, the distress inflicted by regular droughts, floods and crop failures will be reduced. The statistical offices should carefully collect data on such matters to assess the longer term impact of policy.
Use of all village commons has to be under the control of the local panchayat government, and its collective use should be decided by secret voting of members. Employment guarantee schemes have to be extended and oriented to a significant extent for this purpose of creating and improving village commons. However, its possible positive or negative spillovers to private land also have to be considered by the panchayat with mutual measures of local taxes and subsidies. Those enjoying external economies or suffering diseconomies will have to be represented in the local decision-making bodies. Democracy means allowing dissent, and inter- as well as intra-class conflict of interests amongst peasants is almost inevitable. The extent to which it can be resolved at the panchayat level will be a crucial criterion of judging success, but individuals will also have access to legal remedies. Regular elections are a necessary but not a sufficient condition antidote to authoritarianism, even at the gram sabha or panchayat level. This is the reason, as an integral part of evaluating the success of the system, a separate record must be kept of how many people dissented to each major project.
[T]he present government’s generous five-year package of $24 billion in production-linked incentives has created less than 2 lakh jobs, and has not even scratched the surface of the grim reality of educated unemployment…
A main source of increase in land productivity will come from better use of unemployed and underemployed labour. This can only be achieved through collective use of local labour in a more decentralised set-up with greater administrative and fiscal autonomy to the locally elected governments (the panchayat and gram sabha). The provision already exists, but mostly not used for panchayats in the Constitution.
Labour mobilisation on a massive scale for development of the countryside was one of the most spectacularly successful features of post-revolution China, and also Vietnam. It was centrally planned and executed in a fairly ruthless authoritarian manner for very large irrigation and infrastructure projects. It was successful in many cases, but was based on compulsion rather than voluntary participation. In the Indian case, this cannot work, but widespread unemployment implicitly creates a similar situation in some ways. Poverty compels poor people to take up work as is available. So long as unemployment and poverty persist on such a large scale, employment guarantee at a fixed minimum wage for a certain number of days per year should be the universal right of all citizens. The need to work under the supervision of the panchayat will ensure that this becomes a self-selection process that would be worthwhile mostly for those who really need such work compelled by poverty. It can be a combination of mutually agreed piecemeal and contractual wage payment depending on the nature of work. Wage payment for work is a better alternative than a minimum income scheme. The latter encourages a patron-client relationship between the state and its citizens without any notion of reciprocity between duty and right. This is also the problem of freebies as election promises. Instead, we must strive towards locally decided work on projects which benefits particularly those who participate in the work.
One innovation balancing private with social benefit is to introduce the idea of using labour from the common pool on individual farms by paying a higher wage by the individual owner of land. But this should be allowed only with the consent of the workers and the panchayat. The danger is a two-tier wage system for public and private work. Therefore, it should be introduced only if enough useful public work is not available.
Wage payment for work is a better alternative than a minimum income scheme. The latter encourages a patron-client relationship between the state and its citizens without any notion of reciprocity between duty and right.
For making a dent into the employment problem on a wider scale, improving land productivity on small farms will have to be complemented by a rapid expansion in the services sector in selected directions. Economically, both small-scale industries and services will be subject to some extent by demand-led expansion created by higher income, particularly of small farms. The expanding demand will usually be less suitable for large corporate industries, which typically cater more to upper class consumers and international markets. This results in a natural segmentation of many product markets, and will provide greater impetus for small-scale industry and services expansion when supported by supplementary credit and marketing policies from the government. The government’s policies (like demonetisation and the present cumbersome GST regime) are often inimical to small-scale industries, and have missed the point entirely in this respect. For instance, the present government’s generous five-year package of $24 billion in production-linked incentives has created less than 2 lakh jobs, and has not even scratched the surface of the grim reality of educated unemployment, when some 40 million are currently enrolled in higher education.
The organised, formal sector presently shifts business to the informal sector which causes loss to government revenue through regulation and tax arbitrage. This impetus to the informal sector through subcontracting for tax and regulation avoidance should be handled not by making the regulation regime more suffocating for business, but by encouraging product market segmentation for the range of products that the small scale sector can produce for the poor through greater linkage with small scale agricultural production.
[T]he biggest potential for expansion of employment probably exists through expansion of basic welfare services like delivery of primary health and education.
In this context, the biggest potential for expansion of employment probably exists through expansion of basic welfare services like delivery of primary health and education. They are not only public goods needed desperately by the poor, but they can become a part of the social wage, rather than simply trying to increase private monetary wages to cover expenditure on those basic items. The panchayats should not only be able to deliver them from their fiscal autonomy, but they should be free to devise a tax based on the user-cost principle to partly finance them. Many amongst the educated unemployed can be absorbed in this way with suitable short-duration training. This can also become a useful criterion to judge the performance of the local government for the next round of budget allocation from the state and central governments in a three-tier constitutional structure based on elections, degrees of financial autonomy and governance.
It bears emphasis that a local election-based decentralised system of decision making with local autonomy of local governments in fiscal matters and project selection, while essential, may be difficult to imagine today. However, we may recall that greater autonomy for the states in the Indian federal structure was also gradually achieved, and needs continuous defending even today against centralisation of power. Only if a large number of panchayats and gram sabhas can earn the support of the poor peasants, the constitutional guarantees for their fiscal and other autonomy will survive to make them vibrant. This is the only path to development open to us without over-centralization of power in the hands of the government, supported by a handful of large corporations.
It will be utterly misleading to think of these sketchy suggestions as a complete blueprint for economic development focused on employment. This is meant to stimulate an alternative imagination, an imagination of development that escapes the trap of corporate-led growth that breeds more and more inequality and environmental degradation. Even worse, in the name of economic development, a dangerous mutualism between authoritarian government and a few favoured large corporations develops. It leaves us with only the empty shell of democracy with all its contents hollowed out.
Our main intellectual effort must be to imagine an alternative that is not only desirable, but also feasible in the short run, and relevant in the present situation. Economic development is a long journey, but even the longest journey must start with a first step. This is only an outline of that first step. With an alternative imagination we can face the real challenge and develop with a credible alternative politics, and an alternative narrative of what we want to achieve. Time is running out for that.
Amit Bhaduri is a former Professor of Clear Fame, University of Pavia, Italy, and a former Emeritus Professor, Jawaharlal Nehru University, New Delhi.
