The recent work of Dani Rodrik and two of us (forthcoming) suggests that the question might be wrongly posed. Development theory and experience have shifted the focus to identifying the underlying characteristics that allow a sector to transform an economy and durably raise living standards. Five such characteristics are:
(i) High level of productivity: A country's standard of living is determined by its level of productivity, so economies cannot be transformed by low productivity sectors.
(ii) Dynamism or rapid productivity growth: Resources that move into a dynamic sector experience rising productivity - known as "convergence" toward the domestic or international productivity frontier. This increases economy-wide productivity. A dynamic sector acts as an "escalator" to higher levels of economy-wide productivity.
(iii) Expansion: To ensure that productivity gains from convergence spread, the converging sector must expand, absorbing resources from the rest of the economy. Otherwise, resources outside of the growing sector will not experience higher productivity growth.
(iv) Alignment with comparative advantage: To ensure that expansion occurs and the benefits of fast-growing sectors are widely shared across the labour force, the skill requirements of the expanding sector should match the skill endowment of the country. In labour-abundant India, the rapidly growing sector should be low-skilled intensive, so that more people can benefit from convergence.
(v) Exportability: Part of the reason that exports are important is that trade serves as a mechanism for technology transfer and learning, which may have spillovers on related industries.
More important is this fact: no country in the post-war period (or indeed before) has consistently grown at eight-plus per cent without rapid export growth. Even the Indian growth boom of the 2000s was associated with rapid growth of exports of goods (about 25 per cent) and services (over 30 per cent). Exports provide a source of relatively unconstrained demand for the expanding sector. Put differently, there is no way to sustain high growth by catering only to domestic markets, no matter how big they are. The fear that the world cannot absorb another China-as-India is exaggerated. This fear was raised about the East Asian countries. It was raised about China. In both cases, absorption into the global economy was successfully achieved.
Agriculture will, of course, remain terribly important in India if not as a transformational activity, then as one which has the potential to inhibit transformation both politically and via its impact on inflation.
Three points about India are worth highlighting.
First, an important distinction must be made between unregistered (or informal) manufacturing and registered manufacturing. The productivity and dynamism of the former is considerably less. Specialising in this sector will not lead to economic transformation.
Second, the real Indian dilemma is that sectors that have high levels of productivity and dynamism - and hence transformational potential - are sectors that: use relatively scarce skilled and semi-skilled labour rather than abundantly supplied unskilled labour; and have not absorbed resources from the rest of the economy. Registered manufacturing, for example, is a very dynamic sector but has not absorbed labour from other sectors over the last 35 years - a phenomenon we have described as "premature non-industrialisation."
Significantly, the average education level of employees in registered manufacturing (as well as other dynamic sectors that have been slow to absorb labour such as financial and telecommunications services) is well above the Indian average. Their benefits cannot spread to most of the labour force, limiting the ability of these sectors to generate inclusive economy-wide growth.
Third, one sector that stands out is construction: it is reasonably dynamic, does not require employees to be highly educated, and has expanded significantly in the last three decades. However, the sector is not tradable, limiting its growth potential.
The significance of the prime minister's 'Make in India' campaign - and the associated policy changes, including easing the cost of doing business, reforming labour and land laws and building world class infrastructure - is to make registered manufacturing more unskilled labour intensive and globally competitive so that India's many unskilled poor can benefit from the sector's transformational power.
The significance of the Skill India campaign is to transform more unskilled labour into the skilled labour category. This will allow a wider section of Indian society to benefit from the 'escalator' of India's dynamic services sector.
'Make in India' and Skill India are complementary to each other. To paraphrase the Chinese economic debates of the Maoist era, perhaps India must "walk on both legs."
Amirapu is a PhD student at Boston University; George is a PhD student at Harvard and an economist at the Ministry of Finance; and Subramanian is chief economic advisor at the Ministry of Finance Part five appeared last Friday