Private consumption, constituting nearly 60% of India’s GDP, remains the cornerstone of economic momentum, powered by rising disposable incomes, digital adoption, and an aspirational middle class. Manufacturing contributes about 17% to GDP with aspirations to reach 25% by 2030.
An NCAER survey shows that Indian firms seemed to have weathered the US tariff storm, for now
Amidst the international uncertainty caused by rapid announcements of US trade policies balanced by rationalisation of GST rates, business sentiments as measured by the NCAER Business Confidence Index (BCI) had eased in the second quarter of 2025-26. The most punitive of the policies was the additional tariffs imposed by the Trump Administration. Although their scarring effect may mitigate over time if India-US are able to come to a trade agreement.
Meanwhile, India’s exports to the US came down by 20.4 per cent between August and September 2025 on a month-on-month (m-o-m) basis (Ministry of Commerce). They came down by 11.9 per cent compared to September 2024 on a year-on-year basis. However, overall India’s exports went up by 6.8 per cent on a y-o-y basis and by 4.2 per cent on a m-0-m basis. The fall in India’s exports to the US did not dent overall Indian exports. So, what did India do right? Indian exporters diverted their exports to other countries.
NCAER survey
Does that mean that sentiments were shaken? Well, appears they were just ‘stirred’. The NCAER’s Business Expectations Survey (BES) for the quarter ending September threw up some interesting data. It was carried out days after the imposition of the additional tariffs by the US and almost coinciding with the GST reform and covered 484 firms across all four regions of the country. Business sentiments are assessed by firms’ perceptions on four key components in the NCAER BES: ‘overall economic conditions to improve in next six months’; ‘financial position of the firms will improve in next six months’; ‘present investment climate is positive’; and ‘present capacity utilisation was close to or above optimal level’.
Among the 484 firms surveyed, 43 per cent of firms were exporting to foreign markets, and only 17 per cent directly exported to the US. The sample was divided into three groups – firms that exported to the US along with other countries; firms that exported to non-US countries; and firms that did not export. The data threw up interesting insights about business sentiments.
Majority of firms see no impact
A large majority of firms, 80 per cent, reported no impact of US tariffs on their business, as did 55 per cent of exporting firms, and 99 per cent of non-exporting firms. However, among the 17 per cent of firms directly exporting to the US, 82 per cent reported a negative impact of tariffs on their business.
In alignment, business sentiments were the most elevated, on all four aforementioned components of BCI, among the firms that exported to non-US countries. On the contrary, sentiments were the most subdued among the firms that exported to the US.
Only 39.8 per cent of firms exporting to the US perceived that the ‘present investment climate’ is positive compared to 65.8 per cent of firms exporting to non-US countries and 55 per cent for all firms. Sixty-one per cent of firms exporting to the US perceived ‘overall economic conditions to improve in the next six months’ compared to overall 67 per cent firms. Fifty-eight per cent of firms exporting to the US perceived that ‘financial position of the firms will improve in next six months’ compared to the overall 63 per cent. Sentiments about ‘present capacity utilisation was close to or above optimal level’ for 97.6 per cent of firms that exported to the US compared to 98.1 per cent for all firms. Further, sentiments did not differ across manufacturing and services firms.
Firms diversify markets
To overcome the impact of US tariffs on their business, the majority of firms exporting to the US planned to diversify their markets, either by exporting existing products to countries other than the US, or by redirecting sale of existing products domestically in India, or reduce profit margins of own goods/services to keep prices low.
The majority of firms, negatively impacted by US tariffs, asked for increase in export subsidies, trade agreements with other countries and enhanced credit availability. On the other hand, the firms called for lowering import duties on inputs/raw materials, GS T rates ( which was already implemented), logistics costs and excise duties on petroleum products.
In the survey outcome, it was not a surprise to find that the impact of Trump’s tariffs was limited to firms exporting to the US. Majority of them also exported to other non-US countries and revealed that their dominant strategy was to divert exports to other countries, besides the domestic market.
More FTAs urged
The resilience of Indian firms and timely action by the Indian government appears to have helped mitigate the fallout from the US tariffs in the short run. However, a long-run approach is also needed, which should address the bleak investment sentiments.
India’s exports increased to the UAE and the UK in September, with whom we have trade agreements. And this is also the request of firms from the government. Concluding more trade agreements, combined with additional domestic reforms, could make Indian firms more competitive in the long run.
Bhandari is a Professor, Dayal and Sahu are Fellows and Urs is an Associate Fellow at NCAER. Views are personal.
Could there be a moratorium on further studies diagnosing what ails African development? The continent has been examined, measured and prescribed to exhaustion – yet the patient deteriorates.
Since 2000, multilateral institutions have catalogued the same barriers: mispriced capital, unsustainable debt, illicit outflows, and governance asymmetries. What Africa lacks is not diagnosis but structural change. The Trevor Manuel G20 Africa Expert Panel Report, released during South Africa’s historic G20 presidency, deserves to become the last word – not because it discovers new pathologies, but because it finally prescribes systemic treatment.
The report reframes Africa’s constraints as a single system of mispricing, debt compression and governance asymmetry. Its proposals for refinancing, collective bargaining and International Monetary Fund quota reform mark the first coordinated attempt to shift power within the international financial architecture.
The Manuel framework
In my view, this report sketches what could become Africa’s first genuine Bretton Woods moment. To be precise, Bretton Woods in 1944 did not merely create institutions, it embedded a new power balance into global architecture. The US and the UK wrote rules reflecting their interests and those rules shaped capital flows for decades.
The Manuel framework proposes something analogous: not adjustment within existing laws, but contestation over who writes them. Whether it succeeds is uncertain, but ambition is structural, not incremental.
The arithmetic of transformation
By 2035, Africa must integrate a $3.4tn intra-continental market under the African Continental Free Trade Area, according to African Union and McKinsey estimates, electrify 600m off-grid citizens and absorb 500m young workers into productive employment. Infrastructure financing requires approximately $130-170bn annually; current flows reach barely half of that. This financing gap is the result not of market malfunction but policy shortcomings reinforced by four interlocking barriers.
Figure 1. Policy failure sustained by four barriers
Barrier, mechanism and annual cost
| Barrier | Mechanism | Annual cost |
| Capital mispricing | African sovereigns pay 200-500 basis points above peers despite infrastructure returns outperforming global benchmarks | $45–60bn |
| Debt compression | Debt service crowds out public investment as demographic pressures mount | $89bn |
| Illicit outflows | Trade misinvoicing, tax evasion, beneficial ownership opacity | $88.6bn |
| Governance distortion | Africa collectively holds approximately 6.5% of IMF voting power (about 5% for Sub-Saharan Africa alone) versus ~47% of active lending programmes; 54 sovereigns versus coordinated creditors | Structural |
Sources: G20 Trevor Manuel Report, United Nations Conference on Trade and Development, International Finance Corporation, IMF
Taken together, these outcomes constitute a systemic equilibrium rather than a collection of discrete imbalances.
Each barrier reinforces the others. Mispriced capital raises borrowing costs; higher costs expand debt burdens; expanding burdens compress fiscal space; compressed space weakens institutions; weakened institutions worsen risk perception, which then further misprices capital.
Elevated premia mechanically increase debt service costs; higher debt service compresses fiscal space and limits public investment; constrained investment slows productivity growth and institutional development; weaker institutional performance feeds back into heightened perceptions of risk; and those perceptions then justify the continuation of elevated premia.
Illicit outflows drain resources that could be used to service debt sustainably. Governance asymmetries ensure restructuring favours coordinated creditors over fragmented debtors.
Previous interventions failed because they isolated variables. The Heavily Indebted Poor Countries Initiative reduced stocks but left flows untouched. The common framework offered rescheduling without stock reduction. Rating reforms promised transparency without accountability. Each treated one symptom while others festered. I have watched this pattern repeat across three decades of debt initiatives.
Breaking the equilibrium
The Manuel framework attacks all four barriers simultaneously – the only pathway capable of shifting the system to a new equilibrium.
A strategy that tackles these barriers individually cannot change outcomes; the system reverts to its previous equilibrium unless multiple levers move together. The proposed Jubilee Fund would be capitalised through options, including limited IMF gold revaluation gains or new special drawing rights allocations, enabling debt stock reduction rather than mere deferrals.
The logic is plain: rescheduling preserves net present value for creditors while extending distress for debtors; only principal reduction restores fiscal space. The Brady Plan proved this in Latin America, where 30-35% haircuts catalysed a decade of expansion.
The Borrowers’ Club is, in my assessment, the most consequential innovation. Embedded within the AU, it would harmonise positions on workouts, rating methodologies and climate-finance access. Rather than 54 states negotiating individually against coordinated creditors – including private holders of approximately 43% of African external debt – the Club would create countervailing coordination.
The AU’s Lomé Declaration of May 2025 laid the groundwork: only four African countries have applied to the common framework and completed resolutions have taken over three years. Coordinated positions carry more credibility than isolated ones and shared analytics narrow information asymmetries.
This structural logic extends to governance, where the distribution of decision-making power determines the boundary conditions for all other reforms. On governance, the report advocates an externally anchored IMF review shaped by strengthened African representation.
Africa holds approximately 6.5% of total IMF voting power (around 5% for sub-Saharan Africa alone) while receiving close to half of the Fund’s loan commitments. The 2024 addition of a 25th executive board chair was a modest step but inadequate to the scale of imbalance. Movement towards meaningful quota reform would signal that architecture can evolve.
The path ahead
Reform windows in multilateral systems open rarely and close quickly. The 2026 calendar is unusually aligned: the Borrowers’ Club convenes in Addis Ababa in February; the African Development Bank issues its first climate-resilient bond in Q2; IMF Board papers on refinancing land in April; the 17th quota review concludes in October.
If these milestones hold, participating economies could redirect 3-4% of gross domestic product from debt service to human capital, infrastructure and climate resilience. This is not a marginal gain.
But none of it is automatic. Creditors must accept losses; institutions must cede authority. With G20 implementation rates averaging around 15%, the burden of collective discipline falls on African governments. History shows how quickly such initiatives fracture when creditors offer bilateral inducements to individual states. Domestic political transitions, exposure to local-currency debt and variation in creditor composition will further complicate attempts at sustained coordination.
The question that remains is whether these proposals can survive the political and incentive pressures that accompany real-world implementation. The Manuel report has clarified the stakes and sketched the path.
Whether Africa consolidates behind this agenda – and whether institutions yield – will determine whether 2026 marks a genuine inflection or another cycle of analysis without consequence.
Udaibir Das is Member of the OMFIF Advisory Council, Visiting Professor at the National Council of Applied Economic Research, Senior Non-Resident Adviser at the Bank of England, Senior Adviser of the International Forum for Sovereign Wealth Funds, and Distinguished Fellow at the Observer Research Foundation-America.
Real-time digital hiring signals, analysed through AI, offer India a powerful new way to anticipate and close emerging skill gaps
India’s debate on skills often concentrate on curriculum reforms, institutional capacity, and the widening gap between education and employer needs. However, a more fundamental challenge concerns our ability to interpret the labour market almost in real time to assess skill shortages (a shortfall in the number of workers required to meet a particular vacancy) and skill gaps.
For years, India has relied on surveys, administrative data, and employer consultations to identify skill shortages. Yet these methods are now insufficient in a labour market transformed by Artificial Intelligence (AI), cloud computing, e-commerce, platform-based work, electric-vehicle manufacturing, digitised services, and rapid technological developments.
The nature of work is evolving swiftly, with new roles emerging and demand fluctuating monthly, making traditional tracking methods inadequate. Fortunately, India’s digital platforms generate continuous, high-frequency indicators that reveal how skills are changing. When these signals are responsibly analysed using AI and machine learning (ML), they allow skilling systems to become more adaptable, precise, and forward-looking.
Key signals come from major job portals such as Naukri, LinkedIn Jobs, and similar sites. These platforms list millions of job postings with detailed descriptions of the skills employers seek. They offer insights beyond simple vacancies, indicating which technical tools are gaining popularity, which job categories are expanding, how salary ranges are evolving, where companies are flexible on experience, and which cities are becoming hiring hubs.
Advanced natural-language processing (NLP) and ML models can analyse these large datasets to identify emerging roles, regional hiring trends, and early indications of skill shortages. These AI tools can categorise job descriptions, group similar roles, analyse skill overlaps, and track the spread of new technologies across sectors.
Professional networking platforms like LinkedIn offer another essential layer of workforce insight. Signals such as recruiter posts, urgent job openings, referral requests, emerging certification trends, changes in candidate search locations, and surges in skill endorsements collectively form a behavioural map of labour-market dynamics. Using AI-based pattern recognition, these indicators reveal how long job openings remain unfilled, where mid-career shortages are intensifying, and how occupational shifts occur in real time. These insights originate from genuine professional activity and serve as the digital fingerprints of the labour market.
India’s platform economy broadens this landscape beyond white-collar jobs, providing insights into emerging economic hotspots across other sectors. Platforms such as Urban Company, Swiggy, Zomato, Ola, Uber, and Upwork offer ongoing indicators of demand for technicians, drivers, delivery workers, freelancers, electricians, and repair experts. AI and ML tools can analyse these behavioural data to identify shifts in consumption, logistical issues, and technological adoption — whether it’s the growth of rooftop solar, the rise of home automation, or the expansion of EV charging stations.
Image-recognition models can also analyse geotagged visual data from public sources such as construction activity, warehouse occupancy, EV infrastructure growth, or retail footfall, to gauge local economic activity and forecast future workforce demand. Additionally, ed-tech platforms with millions of monthly learners reveal which skills young Indians view as future-ready. ML models can identify growing interest in areas such as cloud security, product management, data science, and digital marketing in Tier-II and Tier-III cities, revealing emerging talent pools before they appear in corporate datasets.
AI/ML-based methods of assessing and forecasting demand for workers may not apply to more traditional sectors or to sectors where firms hire through networks rather than public job postings. Nor do they fully address skill gaps, i.e. cases in which workers have the appropriate educational background but lack the requisite skills.
However, alternative data need not substitute traditional surveys and employer consultations; instead, it complements and reinforces existing trends. For example, AI models may detect rising demand for embedded-systems engineers in a specific area. This could prompt targeted surveys or industry discussions to understand the underlying causes.
Persistent recruitment difficulties for mid-career data architects identified on LinkedIn may encourage Sector Skill Councils to redesign upskilling initiatives. An increase in cybersecurity course enrolments might indicate regional ambitions for digital roles, prompting institutions to expand relevant offerings. These indicators provide evidence-based guidance for decision-making rather than relying solely on intuition.
India’s research community has already begun adopting this integrated approach. The MSDE-National Council of Applied Economic Research (NCAER), in its national skill-gap study across high-growth sectors, used alternative data sources alongside surveys and firm interviews. This mixed approach—enhanced by AI-driven trend analysis—recognises that India’s labour market is too fluid for traditional tools alone. It demonstrates how integrating conventional data with digital signals offers a more comprehensive, current, and precise picture of industry shifts.
For example, using Naukri.com date for nine cities in 2024, we found that most advertised jobs were in data and cloud roles, with over 70 per cent in the BFSI sector. Different cities exhibit different patterns. Bengaluru sees strong demand for software development, while Delhi shows high demand for both data-and-cloud roles and software development. Both cities, however, show the greatest need for workers with 2–5 years of experience or more than five years.
Instead of waiting for labour-market trends to materialise, India can now forecast them in advance. Through AI-based analytics, skill-adjacency modelling, and predictive labour-market forecasting, universities can update their curricula more rapidly; Sector Skill Councils can revise occupational standards more regularly; states can identify emerging talent hotspots early; and employers can build workforce pipelines before shortages become critical.
Young Indians also stand to gain, as real-time insights into growing occupations, wage patterns, and regional demand trends can help them make better- informed decisions about skills and career pathways.
The skilling ecosystem can be transformed by responsibly utilising digital signals from hiring platforms, labour-market activity, and workforce behaviour, all while adhering to strict privacy standards under the DPDP Act. With world-class digital infrastructure, abundant talent, and increasing momentum, India can build a skilling system that is far more adaptive and insight-driven. By harnessing digital hiring data, platform behaviour, AI-enabled analytics, ML-based trend recognition, and image-driven economic indicators, India can create a dynamic, future-oriented workforce strategy aligned with its rapidly evolving economy. Such an approach embodies the vision of ‘Viksit Bharat’, where opportunity and capabilities rise together, and every region has the skills foundation to participate fully in India’s growth journey.
Views are personal.
Though the incidence of son preference in Indian families has fallen, there is still little preference for daughters.
Instead of asking parents, just count their children to get a sense of what India genuinely values. From 1.33 males for every girl at the first birth to 1.43 at the second and 1.51 at the third, the sex ratio of a family’s last child becomes increasingly skewed over NFHS-5.
This is deliberate; it’s not a coincidence. This nation has a long tradition of waiting for a son before allowing fertility to stop. And no phrase captures the consequences of this behaviour more sharply than the one introduced in the Economic Survey 2017-18: “21 million unwanted girls.” The term may have disappeared from headlines, but the behaviour that produced it remains firmly in place. The last child still tells the truth.
This insight becomes more evident when we analyse the evolution of attitudes over time. India today articulates a more confident stance regarding equality between sons and daughters. Survey responses seem promising. Parents are significantly less inclined than three decades ago to explicitly express a preference for males.
Girl reluctance
However, scarcely anyone articulates a preference for daughters. And asNFHS-5 demonstrates, “equal preference” frequently conceals a persistent reluctance to terminate fertility with a female child. The transition is primarily verbal rather than behavioural.
The NFHS trend over 30 years captures this gap between changing language and unchanging lineage expectations. Son preference in attitudes has fallen; daughter preference has barely risen. The supposedly neutral “equal preference” category has expanded instead, but it has not translated into gender-neutral stopping behaviour.
This is why the sex of the last-born child remains the most honest indicator of what families actually want. If India had moved beyond son preference, the sex ratio of the last child would hover near natural levels. Instead, as birth order rises, the skew deepens. Families rarely stop after a daughter, but almost always stop after a son. In demographic terms, this pattern is called “son-biased stopping”. In human terms, it means daughters are rarely the child that completes a family.
Story of unwanted girls
These are not merely numbers. They represent the structural forces that create “unwanted girls.” A girl becomes “unwanted” not because her family dislikes her, but because she was born in a household that continued childbearing to produce a son. She enters a family already larger than intended, already stretched in terms of income, nutrition, attention, and care. Research consistently shows that last-born girls in son-seeking families receive fewer vaccinations, are breastfed for shorter durations, have lower nutritional outcomes, and drop out of school earlier. It is rarely deliberate discrimination; it is the silent arithmetic of scarcity in households that exceed their desired size.
The enduring nature of this pattern also accounts for why development alone has not addressed son preference. Education enhances attitudes but does not entirely eliminate patrilineal expectations. Increasing in-comes generate opportunities but do not undermine the norms that as-sociate family name, rituals, inheritance, and old-age security with sons.
Even in States characterised by high levels of human development —such as Punjab, Haryana — the sex ratio of the most recent infant re-mains markedly skewed in favour of males.
North-East outlier
The one region that consistently breaks this pattern is the North-East. In Meghalaya, where inheritance flows through daughters in its matrilineal system, daughter preference exceeds son preference. Nagaland, Mizoram, Manipur, and Sikkim show milder versions of this trend.
Where daughters remain within the natal family — legally, ritually, and economically— fertility patterns become gender-neutral. In contrast, across much of India, daughters are still imagined as leaving the family, and sons as carrying it forward. Culture, not wealth, determines whether a daughter can be the last child.
Little behavioural change
This is why the argument that the issue is outdated misses the point. The terminology may have entered the public lexicon in 2018, butNFHS-5 proves that the underlying behaviour remains alive. The Economic Survey offered a vocabulary — unwanted girls, son meta-preference, last-born indicators. The latest data shows these are not historical curiosities, but contemporary realities. The story is not old; the fail-ure to address it is.
India’s policy approach has long focused on preventing sex-selective abortion. While this legal restriction was essential to address the enormous crisis of “missing girls,” it also shifted the problem rather than solving it. When families could determine the sex of the foetus, the girlswho were born were, at the very least, wanted.
Once the law made that impossible — as it rightly did — the same underlying preference reappeared in a different form. Parents who cannot choose the sex of a child before birth often pursue the same outcome after birth by continuing childbearing. The result is fewer “missing girls” but more “unwanted girls” —daughters born not because the family chose them, but because the family was still waiting for a son.
Limits of law
The site of discrimination has moved from the prenatal clinic to the everyday home, revealing that laws can curb a practice but cannot, by themselves, transform the norms that drive it.
Correcting this requires more than incremental reform; it demands a structural recalibration of how families, communities and institutions value daughters. Families need reliable, universal old-age support that does not force them to treat sons as insurance policies. Inheritance and property rights for daughters must be implemented with intent — enforced in practice, not admired only in legislation.
Women must have real agency in decisions about if and when to have children, because fertility patterns will not change as long as patriarchal expectations decide the acceptable end-point of a family. And India must nurture a cultural shift in which lineage, obligation and identity are no longer imagined as the exclusive domain of boys.
Until these deeper transformations take root, India’s demographic data will continue to repeat the same quiet message. Fertility may fall, education may rise, and attitudes may soften, but the sex of the last-born child will keep exposing what families cannot say aloud. The day a daughter can be the final child because the family feels complete — not because its patience has worn thin — will be the day India moves be-yond its hidden hierarchy of births. Until then, the last child will continue to tell the truth.
Deka is Fellow at NCAER; Das is Assistant Professor at IEG, Delhi. Views are personal.