Structurally strong, cyclically squeezed — rates, fiscal, sector policy, flows, and two trade ideas with hedges.

3.1 Rates & the FCNR(B) story

RBI opened a special FCNR(B) dollar-swap window (8 Jun–30 Sep 2026), letting NRIs earn 5.5–7% tax-free on 3–5yr USD deposits, with RBI swapping the dollars for rupees. Sized for ~$50bn — banks brought in ~$133bn. Mechanism: NRI deposits dollars → bank swaps with RBI for rupees at a forward rate → RBI's reserves get a shot in the arm (they'd fallen from a $728bn Feb peak to ~$682bn) → RBI can lean against rupee weakness without burning existing reserves.

The catch: $133bn swapped into rupees is a large liquidity injection. To keep it from being inflationary, RBI has had to drain it back out — ~₹1 trillion of OMO bond sales through September. That selling, stacked on the oil-driven yield sell-off, is why the 10Y G-sec just broke 7% for the first time in three months.

3.2 Government borrowing & the crowding-out channel today's news

The Centre just cut its FY27 gross market borrowing to ₹15.99 lakh crore, down from the Budget Estimate of ₹17.20 lakh crore — a ₹1.2 lakh crore reduction. H2 FY27 borrowing via dated securities: ₹7.86 lakh crore across 23 weekly auctions (including ₹15,000cr sovereign green bonds), with a higher share of ultra-long paper (15Y up from 14.5% to 17.6% of issuance, plus more 30/40/50Y). Driver: stronger-than-budgeted revenue — tax revenue of ₹8.45 lakh crore and non-tax revenue of ₹4.23 lakh crore in the first four months of FY27, lifted by a large RBI dividend transfer (₹2.87 lakh crore, up from ₹2.69 lakh crore) — plus a bond-switch operation that cut the FY27 repayment need to ₹4.36 trillion from a budgeted ₹5.47 trillion. FY27 fiscal deficit target: 4.3% of GDP, down from FY26's revised 4.4%.

Why "government borrows less" is good — the mechanism, precisely: government finances its deficit by issuing G-secs, which compete with corporate bonds and bank credit for the same pool of savings. To place a large borrowing programme, it has to offer yields attractive enough to draw that much capital — pulling up the whole rate structure, since G-sec yields are the risk-free base every other loan and bond in the economy is priced off. Higher risk-free yields make private borrowing costlier — directly (corporate bonds priced as a spread over G-secs) and indirectly (banks find risk-free G-secs more attractive than lending to companies, "crowding out" credit that would otherwise go to private capex). A smaller borrowing calendar runs this in reverse — less competition for savings, easier yields, cheaper private financing — which is exactly why bond markets rallied on this news.

Nuance worth having ready: crowding-out is strongest when the economy is near full capacity and savings are scarce. When there's slack — spare savings, soft private credit demand — government borrowing can be absorbed without much yield impact, and public capex can even crowd in private investment (a highway or transmission line unlocking capex around it). So "less borrowing is unambiguously good" is directionally right but not a universal law.

The rupee link — the honest version: the direct classical channel is the twin-deficit idea: a lower fiscal deficit reduces the government's net dis-saving, improves headline fiscal metrics, and lowers the sovereign risk premium — generally currency-supportive through credibility, and through crowding in private investment that eventually supports growth and exports. But the mechanical link is weaker for India than for economies that fund deficits externally, because India's G-sec market is overwhelmingly domestically funded (banks, insurers, EPFO, RBI) — smaller borrowing mainly means less competition for domestic savings, not less external borrowing. There's even a partial offset: less G-sec supply and lower yields make Indian bonds slightly less attractive to yield-seeking FPI debt investors (lower carry) — so on the margin, smaller borrowing could modestly reduce, not increase, FPI debt inflows. Net answer: fiscal consolidation supports the rupee mainly via credibility and growth, not a straight mechanical line — and it's genuinely two-sided once you account for carry.

Trader's point to make unprompted: the G-sec curve has two forces pulling opposite ways right now — RBI selling bonds via OMO (drains liquidity, pushes yields up) against the government issuing fewer bonds this year (less supply, pushes yields down). That tension, not a single clean narrative, is the actual state of the India rates market today.

3.3 Fertilizer & urea — from import dependence to self-reliance

India is the world's largest urea importer, sourcing roughly a quarter of its annual requirement from overseas — which is exactly what turned this year's Strait of Hormuz shock (see §3.4) into a fiscal event rather than a footnote. On 15 July 2026 the Cabinet Committee on Economic Affairs approved the National Investment Policy for Urea 2026 (NIPU-2026), a structural response aimed at cutting that import dependence permanently, not just riding out this year's spike.

• The policy has already drawn ~₹2.04 trillion (~$23bn) in investment proposals — 17 of them, from GAIL, Hindustan Urvarak & Rasayan (HURL), Kribhco and others — covering 8–9 new plants plus expansions of existing capacity.
• Design improvements vs. the old 2012 framework: fixed and variable costs are now separated for transparency; a bounded return-on-equity band (12–16%) replaces open-ended subsidy exposure; and — the detail that matters most to an investor — fixed costs convert into rupee terms after four years at the prevailing exchange rate, which is a direct fix for the FX risk that made the 2012 policy unattractive to new capacity.
• The two-speed story to tell in the room: this year had a cyclical fertilizer chapter (the Hormuz-driven price and subsidy shock, §3.4, now easing) and NIPU-2026 is the structural chapter — a multi-year push to stop India being exposed to exactly that kind of shock again. Being able to separate the cyclical shock from the structural policy response is a noticeably better answer than treating "fertilizer" as one undifferentiated story.

3.4 Inflation — what's driving the 4.82% print

• Weak monsoon — rainfall deficit hit ~42% below normal in June, easing to ~15% by September; on track for the weakest monsoon since 2009. Reservoirs fell to ~28% of capacity, pushing farmers onto groundwater/electric irrigation pumps — the direct link from "bad monsoon" to "higher power demand." Economists estimate a weak monsoon can add up to 100bp to CPI.
• Sugar — India expected a surplus and started exporting; red rot disease hit UP cane fields, cutting production estimates from ~343 LMT to ~306 LMT. India reversed course, now allowing 1 million tonnes of duty-free raw sugar imports (normally 100% duty) through 31 Oct. Retail sugar rose ~16% in a month (₹48.18 → ₹55.70/kg, Jul–Aug).
• Wheat — the opposite story: record output (120.65 MT) and 5-year-high FCI stocks (50.5 MT) let India lift its 4-year wheat export ban in August to support farmgate prices — disinflationary for wheat, doesn't offset food broadly.
• Fertilizer/urea — the acute 2026 shock: the Strait of Hormuz crisis briefly cut India's urea output ~25% and pushed import prices toward $935–959/tonne, nearly doubling the projected subsidy bill toward ~₹3.4 lakh crore; it's since eased (Hormuz reopened, urea back near $415–420/tonne) toward ~₹2.5 lakh crore. See §3.3 for the structural policy response this triggered.

3.5 Exports & import dependence

Gems & jewellery: FY26 exports ~$27.7bn, stable despite a US slowdown, helped by a US–India trade-framework agreement. Composition: cut & polished diamonds 43.9%, gold jewellery 41.0%, silver jewellery 5.3%. India cuts and polishes diamonds it doesn't mine — rough stones are imported (~$700m in June alone) — a pure re-export/value-add trade, acutely sensitive to global demand for polished stones and to rough-diamond supply upstream. Best answer if asked which export sector is most fragile to a global supply-chain shock.

3.6 UPI MDR — the new merchant fee

Correction to flag for yourself first: the new fee applies to P2M (person-to-merchant) UPI payments only — not P2P. Every peer-to-peer UPI transfer stays 100% free regardless of amount; don't say "P2P" in the room.

From 15 October 2026, NPCI's new rule applies a 0.4% MDR (merchant discount rate) to eligible P2M UPI transactions above ₹2,000, capped at ₹300 for transactions of ₹75,000 and above (worked examples cited in reporting: ~₹12 on a ₹3,000 payment, ~₹200 on ₹50,000). Small merchants under the P2PM framework taking up to ₹1 lakh/month via UPI QR remain fully exempt — zero MDR, protecting kirana-scale merchants. Consumers pay nothing either way; the fee sits with the merchant.

Why this belongs in a markets briefing: UPI has run on zero MDR since 2020, with payment companies compensated via a government incentive scheme instead of merchant fees — a policy-subsidised, effectively zero-revenue payments rail. This reopens a genuine fee-revenue line for banks and payment aggregators in the ecosystem for the first time in years. It's also a small but clean example of the same fiscal-discipline thread running through this whole India section (smaller government borrowing, tighter subsidy design in urea) — recouping part of the cost of running a public payments rail from the transactions that use it, rather than an open-ended subsidy.

3.7 NSE's own listing — why the debut was muted

The National Stock Exchange of India listed on 24 September 2026 at ₹1,800 — just ~1% above the IPO's upper price band of ₹1,785 — despite the issue being subscribed 5.71 times. Bidding ran 17–21 September; the offer was entirely an Offer For Sale (OFS) of 12.64 crore shares, raising ₹21,486–22,562 crore, one of the largest IPOs in Indian history.

Your instinct — size and magnitude — is a real part of the answer, but not the whole answer:

• Pure OFS, no fresh capital: every rupee raised went to existing shareholders selling down, not into NSE's own balance sheet — there's no new-growth-capital story to re-rate around at listing, only an ownership change among already-large holders.
• Size: at ~₹21,500cr+, absorbing the stock at the margin needs very deep pockets — strong headline subscription (5.71x) can still coexist with a flat listing once large anchor/institutional allocations are already locked in and the marginal secondary-market buyer is thin.
• Years of prior unlisted trading (plausible, not confirmed): NSE shares had traded informally in India's unlisted/grey market for years during its long-delayed IPO saga — a real possibility is that much of the "this is a rare, monopoly-like asset" re-rating had already happened before the formal listing, leaving less of a pop for listing day itself. Flag this one as your own reasoning, not a cited fact, if asked to defend it.

3.8 Equity flows — why India is lagging

India ran up to expensive multiples (22–25x) into 2022, corrected through 2024–25, and is now cheap-ish (~19.7x) — but still lost the flows. Why: (1) India has almost no seat at the AI/data-centre/semiconductor table driving the US, Taiwan and Korea higher — called an "AI loser" this cycle; (2) higher US rates and a firmer post-Warsh dollar pull capital back to the US and out of "relatively expensive" EM; (3) even the money that came back in Jul–Aug went disproportionately into IPOs/primary issuance rather than secondary buying of existing names — a vote for specific new supply, not for India Inc.'s existing valuations. A strong IPO pipeline despite a soft secondary market fits: promoters and PE/VC sponsors use the window to exit at any workable price, and primary demand holds even where index-level conviction hasn't.

3.9 Trade idea — power sector

3.10 Trade idea — metals

3.11 Gold & silver — India angle

Gold's structural bid (central-bank buying, de-dollarization) plus a weak rupee (import cost in INR terms rises even when the dollar price is flat or falling) is a big part of why gold has been a crowded trade for Indian retail/HNI money — and part of why that money hasn't gone into equities instead. Rupee weakness and dollar gold prices can offset each other, so gold-in-rupees can hold up even while gold-in-dollars corrects, as it is right now (see §7.2).