The missing middle: premature deindustrialisation, twin engines and reindustrialisation

Detail

1. The "missing middle": what the phrase means

• Classic path of structural change: workers move from the primary sector (farming) to the secondary sector (factories), and then to the tertiary sector (services). This is the Class 10 three-sector story.
• India's path was different. Workers and output moved from farming straight into services. The factory stage in the middle stayed thin. This thin middle stage is called the "missing middle".
• Starting point of this era: in 1972-73 about three out of every four workers were in the primary sector (Class 11, Employment chapter). Five decades later, farming still employs the largest group.
• Also called the missing middle: India has many tiny informal firms and a few large firms, but few medium-sized firms. Medium firms are what usually grow into big factory employers. (Firm-size policy is covered in industrial-policy-psu-msme.)

2. Premature deindustrialisation (Dani Rodrik, 2016)

• Deindustrialisation: the share of manufacturing in a country's employment and output falls over time.

• In rich countries this happened after they became rich (for example the US and UK after the 1970s). That is normal.

• Premature deindustrialisation: manufacturing's shares of employment and output peak, then decline, at much lower income levels than in today's advanced economies.

• The result is that factories stop absorbing workers before the country becomes rich.

• Evidence (IMF): countries that began deindustrialising only after 1990 saw manufacturing's employment share peak at just 18.9%, at a per capita income of about $4,273 (constant 1990 dollars). That income level is far below the level at which developed countries peaked [7].

• Why it happens:
• Technology: to win in global markets, factories now need machines and skilled workers. Modern production is capital-intensive and skill-intensive (it uses more machines and trained workers, and fewer ordinary workers) [7].
• Competition: China and East Asia already occupy the low-cost factory space.

• Cheaper imports after trade liberalisation squeeze domestic factories in late-developing countries.

• The danger: workers stay stuck in agriculture or move into low-productivity services such as street vending and petty trade. They miss the growth phase that factories gave East Asia [7].

3. India's numbers

• Output share: manufacturing has stayed at about 15-17% of GVA since the 1990s.
• GVA (Gross Value Added) = value of output − value of intermediate inputs. It measures what a sector itself adds.
• Latest official figure: manufacturing's share of GVA was 17.2% (2013-14) and 17.5% (2023-24). It is almost flat [2][3].

• Over the same period, real manufacturing GVA (at constant prices) grew from ₹15.60 lakh crore to ₹28.25 lakh crore [2][3]. Trap: manufacturing output grew fast in absolute terms, but its share did not rise, because services grew just as fast.

• Employment share: manufacturing employs about 11-12% of workers.

• PLFS: manufacturing's share of workers fell from 12.1% (2017-18) to 11.4% (2023-24) [6].

• PLFS (Periodic Labour Force Survey) is MoSPI's (NSO's) main survey of jobs and unemployment. It began in 2017-18.

• Agriculture moved the wrong way: farming's share of workers rose from 44.1% (2017-18) to 46.1% (2023-24) [6].

• The share of women workers in agriculture rose from 57.0% to 64.4% over the same period [6].

• This is reverse structural change: the share of workers in farming went up, not down.

• Comparison: manufacturing shares in East Asia (South Korea, Taiwan, China) peaked at 25-30%.

4. Worked example: why the shift out of farming matters

• Relative labour productivity of a sector = (sector's share of output) ÷ (sector's share of workers).

• A value above 1 means a worker in that sector produces more than the average worker.

• Manufacturing (India): 17.5% of GVA ÷ 11.4% of workers ≈ 1.5. A factory worker produces about 1.5 times the national average.

• Illustrative farm case: suppose farming makes about 18% of GVA with 46% of workers. Then 18 ÷ 46 ≈ 0.4. A farm worker produces about 0.4 times the average.
• Gain from moving one worker: a worker who moves from the farm to a factory raises output by about 3.75 times (1.5 ÷ 0.4). This reallocation gain is the growth that the missing middle gives up.

5. Why manufacturing matters

• Absorbs low-skill labour at scale. Garment, footwear and assembly lines can train a farm worker in weeks. This is exactly what India's surplus farm workers need.

• Disguised unemployment (more people working on a farm than are needed, so removing some would not cut output) means many farm workers can leave without farm output falling.

• Productivity convergence (poor countries' productivity catching up with rich countries' levels):

• Factory productivity tends to catch up with global levels, even in poor countries.

• This happens because machines and methods can be copied, and exports force firms to meet world standards.

• Tradable: factory goods can be sold to world markets.

• So growth is not limited by demand at home. A haircut must be sold locally, but a phone can be sold anywhere.

• Linkages: factories buy inputs from, and sell outputs to, many other sectors. This creates extra jobs in transport, trade and finance. (This is the interdependence of sectors in Class 6 and Class 10, for example the AMUL dairy chain.)

• Services cannot do this job alone. Skill-intensive services such as IT and finance cannot absorb the roughly 46% of workers still in farming, because those workers lack the education these jobs need.

6. Contrast with NCERT

• Class 11, Indian Economy 1950-1990, treats industry's share of GDP rising from 13% (1950-51) to 24.6% (1990-91) as "an important indicator of development."
• Since the 1990s this rise has stalled. Manufacturing is stuck in the 15-17% GVA band.
• Exam angle: NCERT's planning era judged success by how fast industry's share grew. By that same test, the period after 1991 falls short on manufacturing.

7. Counter-view: services as a growth escalator

• Argument: modern services are now tradable and productive, like manufacturing once was.
• IT and IT-enabled services, GCCs (Global Capability Centres, offices where multinational firms run their own technology, finance or R&D work from India), and digital services all earn export income.

• Their productivity can also converge with global levels.

• Limit:

• These services need educated workers.
• They create few jobs for the low-skilled.

• So they can raise GDP, but they cannot empty the farm sector.

• Middle view: services can be one engine of growth, but they need a second, labour-absorbing engine next to them. This leads to the "twin engines" idea below.

8. Policy response: twin engines of growth

• Twin engines of growth: growth powered jointly by manufacturing and services, not by one engine alone.
• Services bring skills, exports and high productivity.

• Manufacturing brings mass jobs for low-skill workers.

• Second meaning (a common trap): the phrase is also used for public investment plus private investment growing together.

• Example: government spending on roads and ports "crowds in" private factories. Crowding in means public spending makes private investment more attractive.

9. Reindustrialisation

• Reindustrialisation: a policy-driven revival of factories after offshoring (firms moving production abroad) or deindustrialisation.
• Global examples:
• US CHIPS and Science Act and Inflation Reduction Act (both 2022). These give subsidies for chips and clean-energy manufacturing.
• EU Chips Act and Net-Zero Industry Act.
• China+1 (firms adding a second production base outside China) and "de-risking" supply chains (cutting dependence on a single country without fully breaking ties).

• The global side is covered in globalisation-mnc.

• Why it matters for India: supply chains leaving China open a window for India. But rich countries are now also paying subsidies to bring factories home, so India faces more competition for them.

10. India's instruments

Instrument	Year	Key point
National Manufacturing Policy	2011	Target of 25% of GDP by 2022 was missed. The share stayed near 17% (17.5% of GVA in 2023-24) [2][3]
Make in India	2014	Aims to attract investment and improve ease of doing business
PLI schemes	2020-21	14 sectors, outlay about ₹1.97 lakh crore [3]
India Semiconductor Mission	2021	Chip and display fabs; ATMP/OSAT (units that assemble, test and package chips)
National Manufacturing Mission	Budget 2025-26	Covers small, medium and large industry to further Make in India [4]

• PLI (Production Linked Incentive): the government pays firms a percentage of their extra sales (sales above a base year) made from products produced in India.
• The subsidy is paid only if production actually rises, so it is linked to output, not to investment.
• Worked example: a firm's base-year sales are ₹1,000 crore and it sells ₹1,500 crore this year. The extra sales are ₹500 crore. At a 5% incentive rate it gets ₹25 crore.
• Sectors (14): mobile phones and electronics, IT hardware, pharmaceuticals, bulk drugs, medical devices, telecom products, food processing, white goods (ACs, LEDs), automobiles and auto parts, specialty steel, textiles, drones, solar PV modules and advanced chemistry cell batteries [3].
• Results by 31 March 2026: actual investment of over ₹2.40 lakh crore. Employment of over 14.15 lakh (direct and indirect) [2].
• Production and sales of over ₹16.5 lakh crore. Exports of over ₹15.2 lakh crore [2][3].
• Examples: capacity of about 55,000 MT set up for 26 critical APIs (bulk drugs, the key ingredients of medicines). This cuts India's dependence on imports [3].
• Compressor capacity (white goods) rose from 1 million units (2021) to 10 million units (2025-26) [3].

• Scale check: 14.15 lakh jobs is small next to the crores of workers in farming. This shows the limit of capital-intensive PLI sectors.

• National Manufacturing Mission (Budget 2025-26):

• It gives policy support, execution roadmaps, and a governance and monitoring framework for central ministries and states [4].
• Five focal areas: (1) ease and cost of doing business; (2) a future-ready workforce for in-demand jobs; (3) a vibrant and dynamic MSME sector; (4) availability of technology; (5) quality products [4].
• Clean-tech manufacturing: solar PV cells, EV batteries, motors and controllers, electrolysers, wind turbines, very-high-voltage transmission equipment and grid-scale batteries [4].
• Official material describes the target as raising manufacturing's share of GDP to 25% by 2035 (verify current) [5]. This is the old 2011 target, now pushed forward in time.

11. The jobs route: labour-intensive manufacturing

• Labour-intensive manufacturing means industries that use many workers for each unit of capital:
• textiles and apparel
• leather and footwear
• food processing

• electronics assembly

• Why these: they match India's large pool of low-skill workers, many of them women leaving farm work [6]. They are also export-oriented, which keeps them tradable.

• Constraints:
• rigid labour rules and informal jobs (covered in employment-informal-sector)
• small firm size and credit gaps for MSMEs (covered in industrial-policy-psu-msme)
• costly logistics and power
• weak skills

Prelims Hooks

• Premature deindustrialisation is linked with Dani Rodrik. Manufacturing's shares peak and decline at lower income levels than in advanced economies.
• For countries that deindustrialised after 1990, manufacturing's employment share peaked at about 18.9%, at around $4,273 (1990 dollars) per capita [7].
• Manufacturing's share of GVA: 17.2% (2013-14) → 17.5% (2023-24). It is nearly flat, even though real manufacturing GVA almost doubled [2][3].
• PLFS: manufacturing's share of workers 12.1% (2017-18) → 11.4% (2023-24). Agriculture's share 44.1% → 46.1% (reverse structural change) [6].
• National Manufacturing Policy 2011 targeted 25% of GDP by 2022. The target was missed.
• PLI: 14 sectors, about ₹1.97 lakh crore outlay. The incentive is paid on extra sales over a base year, not on investment. Results by 31 March 2026: about ₹2.40 lakh crore investment and 14.15 lakh jobs [2][3].
• National Manufacturing Mission (Budget 2025-26) covers small, medium and large industries. It has five focal areas and includes clean-tech manufacturing [4].
• Trap: "twin engines of growth" can mean manufacturing + services or public + private investment.
• CHIPS Act and Inflation Reduction Act (2022) are US reindustrialisation laws. The Net-Zero Industry Act belongs to the EU.
• Class 11 NCERT: industry's share of GDP rose from 13% (1950-51) to 24.6% (1990-91).

Mains Points

• Jobs-growth disconnect: India's growth has been led by services. But skill-intensive services cannot absorb the 46.1% of workers in farming [6]. Without labour-intensive manufacturing, growth stays "jobless" and farm incomes stay low. This links to the demographic dividend (GS-III: inclusive growth, employment).
• Is the PLI model enough? PLI has attracted ₹2.40 lakh crore of investment and created 14.15 lakh jobs [2]. But its sectors are mostly capital-intensive, such as electronics, chips and batteries.
• A jobs-first strategy needs apparel, leather, footwear and food processing to scale up.

• It also needs reforms to labour, land, logistics and MSME credit, not only subsidies.

• Services as an escalator versus a manufacturing revival: GCCs and digital exports prove services can be tradable and productive. The best path is twin engines: services for productivity and foreign exchange, manufacturing for mass employment. Skilling is the bridge between the two.

• The global reindustrialisation race: the CHIPS Act, IRA and EU subsidies, together with China+1, create both an opening and more competition for India. The National Manufacturing Mission (clean-tech focus) and the India Semiconductor Mission place India in this race. Success depends on moving beyond final assembly to higher domestic value addition [4].