Strip gold and silver from core — India’s actual inflation is 2 to 2.5%, not 4%

Oil is the biggest. US commercial inventories including the strategic petroleum reserve have fallen roughly 26% below average to near a five-year low — and the speed matters more than the level. After the Russia-Ukraine shock in 2022 it took about 13 months of reserve releases to produce a fall of that size; this time it has happened in six. That depletion is a large part of why oil didn’t spike despite the Strait of Hormuz closing — the cushion is now spent, and inventory eventually has to be refilled. Do not assume low oil persists. Bloomberg commodity index rose 7% MoM; oil crossed $80 on renewed US-Iran escalation.

Global yields rose almost everywhere — Germany 35 bp, UK 29, Brazil 44 — with only China falling on its growth slowdown. Unusual reason: heavy bond issuance from hyperscalers is crowding out sovereign borrowing rather than the other way round, because the pool of capital is finite.

Trade deficit remains elevated — about $28 billion in May and $30 billion in June — driven by non-oil non-gold imports of electronics and capital goods. Once the FCNR and ECB windows close, that becomes the live question for the rupee.

Monsoon is fine in aggregate but risky in distribution. North-east and east India are at a 30% deficit, southern India at 23% — that’s where regional price spikes come from. Watch distribution, not the national average.

The FCNR window has already brought in about $37 billion against real scepticism that the flows would come at all. Historically 55–60% of such flows arrive in the final month — so at the current run rate they can comfortably cross $80 billion, with another $15–20 billion expected through the ECB window that runs to December. Call it $90–100 billion in total, leaving $50–60 billion still to arrive.

There’s a good reason the market hasn’t fully noticed. Liquidity has not visibly improved yet because the RBI delivered roughly $11 billion against its forward book — cutting its one-month short position from $20 billion in May to $9 billion in June — and intervened in the currency market through July. That’s a timing lag, not an absence. With delivery done and intervention easing, the liquidity should now start to show.

The 1 to 3 year part of the curve is the call — and it is a liquidity call rather than a rate call. A monetary policy pause combined with $50–60 billion of incoming flows is precisely the setup that anchors the short end.

As those flows reach the banking system, banks need to issue fewer CDs — and the short end should see some rally on that supply reduction alone. This is the mechanism to trade, not the rate cycle.

Avoid the long end. It carries global yield sensitivity, currency volatility from geopolitics, and the expectation of eventual rate hikes — risks that are more pronounced at the long end than at the short end.

UTI expects the RBI to hike 50 to 75 basis points later in this cycle — but the bond market is already pricing around 60 basis points. Much of the expected hike is already discounted.

What matters is not whether the hike happens; it’s whether it exceeds what is already discounted. Don’t let the hike headline frighten you out of the trade — the trade is the liquidity mechanism at the short end, and that survives a hike that lands where the market already expects.

The Fed is neither hawkish nor dovish. It is waiting. Three FOMC members voted for a hike — an incrementally hawkish shift — and the new chair has deliberately declined to offer forward guidance, wanting the market to form its own view. Five internal committees are expected to present findings by December. UTI’s read: base effects and AI-related productivity gains slow inflation enough that the Fed may not need to hike materially — a less aggressive path than the market currently prices, and a genuine source of positive surprise.

Bloomberg deferral changes very little. Foreign holdings of Indian government bonds are still less than 4%, FPI flows into Indian bonds have exceeded $6 billion since June (some of it in anticipation of inclusion), and there were no outflows even after the deferral was announced. This market is not hostage to foreign participation. Inclusion, when it comes, is a bonus rather than the thesis.

Growth is holding up. IIP at 7%, GST e-way bill generation up 14.5% by volume, credit growth at 17.5% with industry at 19.2%, services at 21% and agriculture at 16%. Monsoon recovered from a 40% June deficit — the fifth driest June since 1901 — to roughly 10% below normal, with sowing only 3.6% behind last year. Indian bonds proved their resilience: yields rose just 9 bp against 20–40 elsewhere, and are back to 6.75% despite the Bloomberg deferral.

Match the instrument to the horizon. Three to twelve months: money market or low duration. Twelve months to two years: short-term or corporate bond. Beyond two years: income plus arbitrage, for the tax treatment. That laddering matters more than any single rate view.