Cash, Patience and the Apocalypse Trade

I’m watching one date on the calendar more closely than anything else right now: August 27–29, when Fed Chair Kevin Warsh delivers his first Jackson Hole address since taking over from Jerome Powell in May. A slight hint of hawkishness there could force a serious rethink for a lot of investors who, in my view, are still positioned for continuity — an accommodative Fed and an unbroken AI/data-center trade. It’s worth remembering that Warsh has already made a point of withholding forward guidance and that his committee has flagged the possibility of a hike, not a cut, to counter the inflation spike coming out of the West Asia conflict. That is not the backdrop a market priced for calm usually gets.

One data point that sums up how one-sided this positioning has become: the most recent round of large-fund portfolio disclosures showed a decisive pivot deeper into AI infrastructure — data-center operators, cloud capex plays, and chip-adjacent names — even as some of last cycle’s semiconductor favorites were being sold down. When funds of that size are still adding to a single theme this late in the cycle, it tells me positioning risk, not conviction, is doing a lot of the buying.

Zoom out, and the bond market is telling a simpler story: something has to give. Either asset prices crash and deflation sets in, or a large, coordinated QE program gets launched to buy bonds back down. It’s worth remembering that as recently as five years ago, close to $18 trillion of developed-market long-dated government debt traded at a negative yield. We have travelled a very long way from that world.

​The 30-year US Treasury yield touched 5.31% this month — its highest level since June 2003. Japan’s 30-year is at a historical high of its own. A huge cohort of fund managers across the US, Europe, and Japan have simply never operated in a 4.0–5.5% long-yield world, and by my base case, they will not like what the history books tell them about what happens next. Savers, endowments, pension funds, and insurers are all searching for a light at the end of the tunnel. The argument that high inflation lowers the real cost of that debt offers savers little comfort — though fiscal policymakers will certainly take what relief they can get. Financing the fiscal profligacy unleashed since Covid was always going to get harder from here; it is now becoming a genuine strain.

The global financial crisis was met by a coordinated G-20 response that helped prevent a repeat of 1929. I don’t see that same coordination available today. The current state of US relations with the rest of the G-20 is not exactly fertile ground for a joint monetary-and-fiscal rescue, should markets freeze up again.

The optimist’s scenario

By my read, a genuinely benign outcome from here would require all five of the following — and that’s a demanding list:

·         an immediate cessation of hostilities in West Asia, with a full US withdrawal from the Strait of Hormuz and a resulting fall in energy prices;

·         an end to the recurring tariff uncertainty out of the US, freeing up global trade;

·         a meaningful, coordinated bond-buying program (QE) from the Fed, ECB, and BoJ together to bring yields down;

·         a negotiated settlement between Russia and Ukraine, easing the tension in Europe; and

·         an orderly, controlled unwind of the overheated AI/data-center trade.

My base case

A more realistic outcome, in my view, looks like this:

·         a material correction in asset prices, led by a disorderly unwind in the AI/data-center trade, followed by a fragmented (not coordinated) QE response once the damage is done;

·         continued stalemate in both West Asia and Ukraine; and

·         elevated energy prices — Brent averaging around $85/bbl — for at least the next couple of quarters, before demand destruction and alternative shipping routes ease the supply constraint. That figure isn’t a contrarian call on my part; it’s roughly where the EIA’s own outlook already sits, which only adds to my conviction here.

Notice the sequencing in my base case: I’m not expecting central banks to ease into live, war-driven inflation out of choice. I expect them to be forced into it — a crisis-triggered QE that follows a disorderly correction, not one that pre-empts it. That’s an important distinction, because it means the path to easier money likely runs through a genuine drawdown first, not around one.

How to navigate through this

A disorderly correction in US/global AI names would hit Indian equities too. My preference into this kind of setup remains a larger allocation to cash and short-duration debt.

My best trade for the next six months is unglamorous: hold cash, be patient, and go all-in only when the sharp correction actually arrives and the world around you looks and feels like the apocalypse.

The price tag not to be ignored

Over the past week I sat with a stack of very different headlines side by side - a proposed cash payout to women in Uttar Pradesh, an old World Bank note on how Indian states balance their books, a parliamentary panel’s numbers on school dropouts, and a district-level count of doctors in Bihar and UP. Read separately, they are four unrelated stories. Read together, they are one story – sustainability of growth.

A season of generosity

Uttar Pradesh’s government is reportedly weighing a cash payout of up to Rs 50,000 for women in the state, timed just ahead of the 2027 assembly election. The principal opposition, the Samajwadi Party, has countered with a promise of Rs 40,000. This is not an isolated UP story. Over the past three years, the number of states running large unconditional cash transfer schemes for women has gone from two to twelve, and by my reading of the available data, these twelve states will together spend around Rs 1.68 lakh crore on such schemes in 2025-26 alone - roughly 6% of their combined revenue expenditure. Add farmer and youth-linked transfers, and some estimates for women-focused transfers alone run closer to Rs 2 lakh crore across fourteen states.

None of this makes any single scheme wrong on its own terms. Direct transfers can be an efficient, leakage-resistant way to put money in the hands of households that need it, and I don’t doubt many families are genuinely better off for it. My concern, as an investor rather than as a commentator on any party’s politics, is arithmetic, not intent.

Where the money is going

A quick snapshot of what's on the table this election cycle

 ​

​

Six of the twelve states running these schemes are already sitting on a revenue deficit in 2025-26. Strip out the cash transfer spending, and several of them - Karnataka and Madhya Pradesh among them - would show a revenue surplus instead. That gap is not academic. A revenue deficit means a state is borrowing to fund its day-to-day running costs, not its future.

The capex we’re quietly deferring

Zoom out to the national picture and the aggregate numbers still look reasonably disciplined - states’ combined gross fiscal deficit is budgeted at about 3.3% of GDP for 2025-26, broadly where it has held for two years, and outstanding state debt has actually eased slightly as a share of GDP. On the surface, fiscal prudence looks intact.

But the composition worries me more than the headline number. Over the past decade, a large share of the adjustment states have made to hit their deficit targets has come from squeezing capital spending - roads, power, water, hospitals, schools - rather than from trimming recurrent, non-development costs or from mobilizing more revenue. Capital investment has repeatedly acted as the shock absorber whenever a state needed to tighten its belt, in richer and poorer states alike. There are genuine signs of recent improvement - the ratio of revenue spending to capital outlay across states has come down from roughly 6.2 in 2015-16 to about 5.0 last year, which is progress. My worry is that this progress is fragile, and a fresh wave of poll-driven transfer commitments is exactly the kind of pressure that reverses it, because cash transfers are far easier to cut from next year’s capex line than from a promise already made to millions of women voters.

The children and the doctors we don’t have

While states debate the next round of transfers, the human capital numbers tell their own story. A parliamentary panel has just flagged that nearly 73% of Indian students drop out before completing higher secondary school. The number of schools shrinks at almost every stage as children get older - from about 9.1 lakh primary schools down to under 91,000 at the higher secondary level - and enrolment falls in step, from 6.18 crore children in primary classes to just 1.64 crore by the time they reach classes 11 and 12.

Forestation targets, a proxy for longer-horizon public investment discipline, have seen only about a fifth of the decade’s goal met.

Healthcare tells a similar story. India ranks 145th out of 194 countries on WHO health parameters, and the shortfall is starkest exactly where the new cash schemes are being rolled out. Against a WHO benchmark of one doctor per 1,000 people, rural India averages one per roughly 11,000; Bihar is close to one per 28,000, and Uttar Pradesh near one per 20,000. By one estimate, India needs another 6 lakh doctors, 20 lakh nurses and 2 lakh dental surgeons just to close the existing gap - a gap Gujarat’s own recent numbers put in sharp relief, with close to 80,000 new cancer cases and over 43,000 deaths from the disease in a single year.

I don’t read any of this as an argument against welfare spending as a category. I read it as evidence that the current mix of state spending is tilted toward transfers that buy loyalty “this election cycle”, at the expense of the schools, hospitals and roads that would have built durable growth over the next one.

Who actually pays for this

The other half of this equation is who is funding it. Personal income tax has now overtaken corporate tax as India’s single largest source of direct tax revenue, with net direct tax collections crossing roughly Rs 24 lakh crore in 2025-26. That burden sits overwhelmingly on the organized, salaried middle class - a narrow base of taxpayers now effectively out-contributing corporate India, while agricultural income of any size remains untaxed and a large informal economy stays largely outside the net. The compliant formal sector - salaried professionals and organized businesses - carries a growing share of the state’s obligations, including the welfare commitments being made on its behalf without much reference to its own capacity to keep paying.

My takeaway for investment strategy

This is not an immediate crisis - state balance sheets remain broadly sustainable and debt ratios are not alarming today. But sustainability is a medium-term question, not a one-year one, and the direction of travel worries me more than the current level. A political economy that finances consumption ahead of capacity, and funds it by leaning harder on a narrow, already-stretched taxpayer base, is not a formula for the kind of broad-based, decade-long growth story that equity markets like to price in.

For portfolio positioning, I continue to prefer businesses and sectors with pricing power and limited dependence on state fiscal transfers to sustain demand - I’d be cautious of consumption plays whose growth story leans heavily on scheme-driven cash rather than organic income growth. I also continue to doubt a case for holding long-duration debt, given that state finances, like the broader macro picture I have written about before, look calmer on the surface than they may prove to be underneath. 

The balloon could pop faster than the market expects

A few weeks ago, I wrote about a group of people standing in a dark room, each holding a pin, none of them knowing whose hand would find the balloon first. I think I can now name one of the pins. It is called a rate hike.

This is not a small, one-line risk. It sits at the intersection of everything I have been tracking this month: India’s stickier-than-expected inflation, America’s own hawkish undertone, a US political calendar that could remove the last source of pressure on the Fed to go easy, and an AI investment cycle that has been priced for perfection. Put these together, and I think the probability of a rate-hike-triggered correction is higher than most portfolios are prepared for — even though I continue to believe that the longer-run destination for global monetary policy is easier money, not tighter.

Inflation is not cooling the way we hoped

India: a stall, not a reversal

July’s CPI print came in at 4.45% year-on-year, above where most estimates had it pegged, and above June’s 4.38%. Food inflation is running hot at 5.5%, durable food inflation has climbed to 5.1%, and even core inflation — stripping out food and fuel — is holding near 3.9%, with the ‘super-core’ reading (core excluding gold and silver) creeping up to 2.5%. None of these numbers is alarming in isolation. Together, they tell me the disinflation trend of the past year has stalled, not reversed, but stalled at a level still uncomfortably close to the upper end of the RBI’s comfort zone.

My own base case is FY27 CPI inflation averaging around 5%, with a genuine risk of topping 5.5% if the monsoon turns uneven in its closing weeks or crude prices find a bid again. On that math, it would not be unreasonable to expect ~50bps of cumulative tightening in 2HFY27 — a meaningful shift from a market that has spent most of this year debating rate cuts, not hikes.

The US: a hawkish Fed the market isn’t pricing in

The view that the Federal Reserve may hike interest rates by 50-75bps points before this year is out is strengthening gradually, against a Fed Funds futures market that is currently pricing in barely 25 basis points of hikes. That is a large gap between what I think is coming and what the market has actually priced in — and gaps of that size tend to close abruptly, not gradually, once the data forces the market to catch up.

In my view, successful policy needs many conditions to align at once — growth, inflation, credibility, fiscal discipline — while failure can come from any single one going wrong. Right now, resilient growth and sticky inflation are both arguing against the rate cuts the market has been hoping for, which is why my own reading of US rates and the dollar stays firmly on the hawkish side for the next two to three quarters, not the dovish one.

The political overlay nobody wants to say out loud

If the Republicans lose ground in this year’s US midterm elections, the Trump administration effectively becomes a lame-duck presidency for its remaining term. A weakened White House has historically had less leverage — political or otherwise — to lean on the Federal Reserve to keep money cheap. Whatever informal pressure exists today to ease policy in an election-conscious environment would diminish sharply once the electoral calculus behind such pressure disappears.

That matters because a Fed chair with a hawkish reputation, freed from midterm-related political pressure, has every incentive to hold the line on rates for longer than the market currently expects — partly on principle, partly to protect their own legacy. The direction of travel over the medium term may still be lower rates, for reasons I laid out a few weeks ago, but the road between here and there could run through more tightening first, not less.

Why a hike could pop the AI balloon faster than anyone expects

This is where the rate-hike story stops being an isolated macro debate and starts becoming a market-structure problem. Ruchir Sharma and Michael Burry, among others, have separately warned that if US rates move higher from here, the unwind of AI-linked positioning could be sharper and faster than currently priced. I think the underlying numbers explain why that warning has teeth. The RIC report by BoFA strategist digs deeper:

●    Capex without proof of productivity. AI capital expenditure is sitting at record highs relative to sales, while economy-wide productivity growth is running at just about 2.2% — barely above the roughly 2% average this economy has managed since 1987. By my own read, AI’s actual contribution to that number is only about 0.1 percentage point a year, a fraction of what the dot-com capex cycle delivered relative to its own spending. That is a very large bet resting on a very thin evidence base.

●    A crowded, one-way trade. Fund manager surveys consistently show semiconductors as the single most crowded trade in the market, cited by well over three-quarters of respondents, even after brief rotations toward defensives. Crowded trades unwind violently precisely because everyone is trying to exit through the same door at the same time.

●    Concentration risk in the index itself. The ten largest companies in the US market now represent more than $24 trillion, or roughly a third of the broad index. When an index is this concentrated, a reversal in mega-cap leadership does not stay contained to a sector — it drags the benchmark, and every fund benchmarked to it, down with it.

●    A discount-rate problem. AI and other long-duration growth stocks are priced on cash flows expected many years out. Higher rates mechanically shrink the present value of those distant cash flows more than they shrink the value of a bank or an energy company earning cash today. A rate hike does not just raise the cost of capital for AI companies — it re-prices the entire long-duration growth complex at once.

None of this means AI itself is a mirage. There are real, if narrow, productivity gains showing up in areas like software development, and history shows that transformative technologies — railroads, fibre-optic cable, shale — usually do create genuine value eventually, even when the first wave of capital that funds them is destroyed along the way. The risk is not that AI fails as a technology. The risk is that AI-linked equity valuations are pricing in a smooth, uninterrupted path to that value creation, and a rate shock is exactly the kind of event that interrupts smooth paths.

Reconciling this with my ‘more QE eventually’ view

Regular readers will remember that I have also argued the opposite-sounding case: that America’s over-levered household sector, its habit of rolling short-dated Treasury debt, and a banking system running low on reserves all point toward more monetary easing eventually — potentially a ‘super QE’ larger than 2008 or 2020. I still believe that. The two views are not actually in conflict; they describe two different points on the same timeline.

“What’s much less certain is what happens in the interval between ‘everyone knows QE is coming’ and ‘QE actually arrives.’ That gap is where currency stress, bond-market air pockets, and credit events tend to live.”

A rate hike in late 2026 or early 2027 would not contradict the eventual-easing thesis — it would be the disorderly middle chapter that forces the eventual easing to happen.

Higher rates now would be deflationary in their immediate effect: they would raise the cost of AI-related capex, slow the investment momentum that has been propping up GDP growth, and puncture the wealth effect that has kept US consumption resilient even as real household finances have deteriorated. That combination — weaker growth, a wounded wealth effect, and credit stress already visible in pockets of leveraged loans and private credit — is precisely the kind of shock that eventually forces a central bank’s hand toward emergency easing. The path there, though, could be considerably more painful.

Strategy

Indian Equity: Cut to underweight; trim froth. Nifty offers only moderate upside on my own math, and valuations are rich, so macro risk (rate hike + global AI shakeout) argues for buying dips, not chasing rallies. Reduce high-beta, high-PE, story-driven names; favor quality, cash-generative large-caps and domestic-demand plays over globally-linked cyclicals.

Global Equity: Cut exposure. Reduce concentration. Prefer diversified international and value-tilted funds over concentrated AI-themed funds.

Debt/Fixed Income: Rising rates hurt long-dated bonds first and hardest. Prefer accrual, short duration and high-quality corporate paper over long gilts.

Cash/Liquid Funds:  Overweight. Keep dry powder to redeploy if the ‘gap’ between hawkish talk and eventual policy easing turns disorderly. Liquid and ultra-short funds, not idle savings accounts.

(Note: I am not considering gold and cryptocurrencies in my strategy as my understanding of these assets is extremely poor. Each investor may consider these as per his\her understanding.)

The bottom line

The probability of a rate hike — in India, and more importantly in the US — is higher today than the market’s own pricing suggests. If that hike arrives while AI-linked valuations remain this stretched and this concentrated, the resulting correction could be faster and sharper than the consensus ‘soft landing, gentle AI cooldown’ narrative currently assumes. I do not think this changes the eventual destination — a more accommodative monetary regime once the strain becomes visible enough to force policymakers’ hands. But the road between here and there is the part investors actually have to live through, and it is worth positioning your portfolio for a bumpier middle chapter: equity that leans toward quality and away from crowded themes, debt that stays short in duration; and adequate dry powder (cash).

As always, I could be wrong about the timing, the trigger, or the severity — markets have a way of staying irrational for longer than any single argument accounts for. But positioning for a plausible, non-trivial risk is not the same as predicting it, and that distinction is the whole point of risk management.

Gen Z isn’t one crowd

Over the past two months, “Gen Z” has moved from a demographic label to the center of India’s political conversation. Politicians, judges, religious leaders and social influencers who had never before troubled themselves with youth slang are suddenly trying to speak it fluently. Everyone wants a piece of Gen Z’s attention, because everyone has just watched what that attention can do.

The trigger was concrete. A leaked paper in this year’s medical entrance exam, the NEET, forced a retest and set off protests that grew nationwide. What began at Jantar Mantar in early June spilled into a march on Parliament by late July, drew in high-profile supporters, and ended - within weeks - in the resignation of the union education minister.

In a political system that rarely bends quickly to street pressure, this was a genuinely unusual outcome, and it has understandably convinced a great many people that Gen Z is now a decisive force in Indian public life.

A comfortable but misleading assumption

My conversations with a range of people over the past few weeks suggest that most of this new enthusiasm rests on a shaky premise: that India’s Gen Z is a single, homogeneous bloc, united by common grievances and common demands. It isn’t. India’s youth are enormous in number and radically different in circumstance, and that difference matters more than the label suggests.

“Gen Z” conventionally covers those born between 1997 and 2012 - roughly the 15-to-29 age bracket today. The Azim Premji University’s State of Working India 2026 report puts this group at about 36.7 crore, or 367 million people - the largest youth population of any country in the world, and close to a quarter of India’s population.

On the report’s own numbers, this is also nearly a third of the country’s entire working-age population.

That scale is precisely why treating them as one voting bloc, one consumer segment, or one protest movement is a mistake.

Four Indias, not one

The people with the greatest economic grievances often have the least ability to sustain political mobilization - and vice versa.

Break the 36.7 crore down by economic reality and the group splits into at least four distinct constituencies, each shaped by a different set of pressures and each with a very different capacity to act on its grievances.

Read the table left to right and a pattern jumps out: the people with the most acute economic grievances have the least ability to sustain a protest, and the people with the most ability to sustain a protest have the least reason to want the system torn down​

​

Why the loudest group is also the smallest

For a young person from an economically constrained household, life runs on a tight sequence: employment, then income, then family security, then any hope of mobility. Whatever the depth of frustration with the system, weeks spent on a protest site carry a real cost when you are preparing for a government exam, working informally, or simply trying to help your family get by.

The established middle class has a different problem. It has already invested years and real money into education, professional qualifications, and career-building, and it has something to lose from prolonged instability. Its anger, when an exam is compromised, can be intense - but its preferred remedy is to fix the system, not overturn it. This group wants exams conducted properly and merit to count, not disorder.

Rural aspirers - roughly 7.5 crore of them - are arguably as affected as anyone by the gap between promise and opportunity, but they are geographically too dispersed to organize at scale.

That leaves the urban aspirer: highly educated relative to the national average, digitally native, aware of global benchmarks, and carrying a strong, often justified, sense that the system has failed to deliver on its own promises. This is not a group protesting for survival. It is a group protesting because it believes it has been denied something it was told it would earn. At an estimated 3.5 crore, it is also the smallest of the four segments - and the one that just forced a sitting minister to resign.

An old echo, correctly placed

There is a historical parallel worth drawing, though it is often misplaced in casual conversation. Older readers will remember how a wave of student radicalism swept through parts of India in the late 1960s and early 1970s, drawing educated, idealistic young people toward Naxalism and organized Left politics. That movement’s roots were not in Delhi’s university campuses - it began in the Naxalbari uprising in West Bengal in 1967, and its most committed student wing formed in Calcutta, around institutions like Presidency College and Jadavpur University, before spreading to pockets of Bihar, Andhra Pradesh and beyond. What is instructive is not the geography but the mechanism: it was educated, urban, relatively privileged youth - not the poorest of the poor - who supplied the ideological energy, precisely because they had the time, literacy and networks to organize. The same mechanism is visible in this year’s protests.

What this means for anyone allocating capital

·         Policy responsiveness has been demonstrated. A government not known for quick concessions to street protest reversed course within weeks. That is a data point on this administration’s sensitivity to organized, visible, urban discontent - relevant to how markets should price policy risk around education, employment and youth-facing reforms going into any electorally sensitive period.

·         The protest constituency and the consumption constituency substantially overlap. Urban aspirers are also disproportionately the segment driving discretionary spending, digital adoption, and white-collar labour supply. Their frustration with job scarcity - not the exam leak itself - is the more durable signal, and it bears on how confidently one should extrapolate India’s consumption and services growth story forward.

·         The demographic dividend has a shelf life. The same report that supplies today’s headline numbers also flags that India’s youth cohort, currently near its peak, is projected to start shrinking within a decade. The window to convert this population into productive employment - and to avoid the kind of frustration now visible on the streets - is narrower than the growth-story narrative sometimes implies.

·         Civil unrest risk is being independently flagged. Verisk Maplecroft’s Civil Unrest Index has rated India among the highest-risk countries this year. That is a signal worth tracking alongside GDP prints and earnings, not dismissing as unrelated to markets.

The unfinished agenda

None of this means today’s urban aspirer-led movement is destined to fade. It has already shown it can move a government. But 3.5 crore people, however visible and well organized, are a fraction of a 36.7 crore youth population. If this energy is to translate into something more durable than a single ministerial resignation, it will need to broaden its agenda well beyond exam integrity to include the very different concerns of rural and economically constrained youth - jobs, wages, migration, and basic security - rather than assuming those groups share its priorities by default.

Until that broadening happens, my own reading is that this remains a powerful but narrow force: capable of extracting concessions on specific, well-defined grievances, but not yet the singular, unified “Gen Z” movement that the current wave of political and corporate courtship seems to be betting on.

Reimagining India’s energy security

Energy security has been one of the central concerns of Indian policy making for decades, and for good reason. Energy is the lifeblood of economic growth, and our inability to produce enough of it domestically has repeatedly shown up as strain on the country’s fiscal position and trade balance. Every time global oil prices spike, our import bill balloons, our current account deficit widens, and the rupee comes under pressure. This is not an abstract macroeconomic point — it eventually finds its way into fuel prices, inflation, and the fiscal room available to the government for other priorities.

Over the last decade, India has made genuine progress in building a renewable energy ecosystem. But when I look at the numbers, our dependence on imported fossil fuels remains overwhelming, and the high energy intensity of our growth model continues to shape and constrain our foreign policy and geopolitical choices — whether it is our relationship with Gulf oil producers, our purchases of discounted Russian crude, or how we navigate sanctions and shipping routes.

The execution gap

Some of India’s energy problem is structural. But a good part of it is about execution — and the data bears this out.

Crude oil production has been flat to declining for years. Domestic output has fallen from around 29.7 million tonnes in 2021-22 to about 28 million tonnes in 2025-26, even as consumption keeps climbing. As a result, our import dependence for crude has climbed to a record high — touching close to 89% in 2025-26, up from 85.5% five years earlier. India is now the world’s third-largest crude importer, and by some estimates is responsible for a quarter of all global oil demand growth over the last two years.

Natural gas tells a similar story of stagnation, followed by a partial recovery. Domestic gas production had been declining or flat for the better part of a decade before deepwater fields in the Krishna-Godavari basin brought some resurgence — output touched about 35 billion cubic meters in 2023, enough to cover roughly half our demand. But growth from here is expected to be modest, while LNG imports are projected to more than double by 2030, cementing our place among the world’s largest LNG importers.

Hydropower has barely moved the needle. Capacity additions have slowed to a crawl, and hydro’s share of our electricity mix has actually declined over the past decade as other sources — coal, solar, wind — have grown faster. In years of poor monsoon, hydro generation has fallen to multi-year lows, exposing how dependent this “renewable” source still is on rainfall.

Coal is the one fuel where India has genuinely scaled up domestic production — crossing the milestone of one billion tonnes in a single year for the first time, with output up by roughly 70-80% over the past decade. And yet we still import 20-25% of our coal requirement, mainly high-quality coking coal for steel-making and better-grade thermal coal that domestic mines cannot fully substitute. Even our biggest domestic energy success story still carries an import tail we haven’t managed to cut.

And then there is solar — the poster child of India’s renewable transition, and also the clearest illustration of “poor execution” in a different sense. India added a record amount of solar capacity in 2025, pushing installed solar capacity past 130 GW and non-fossil capacity past half of our total installed power capacity — a Paris Agreement milestone reached five years ahead of schedule. On paper this looks like a triumph. In practice, a meaningful share of this new capacity has had to be switched off. Grid operators curtailed roughly 2.3 terawatt-hours of solar generation in 2025 alone because the system simply could not absorb it — Rajasthan, our leading solar state, has seen curtailment cross 50% during peak sunlight hours in some months. The core problem is that we built generation capacity far faster than we built storage and grid flexibility: India’s battery storage capacity, at last count, was a fraction of a single gigawatt-hour — negligible compared to the scale of solar we have added. We built the tap before we built the tank.

Beyond Megawatts: What a holistic framework looks like

This brings me to what I think is the deeper issue: India’s energy conversation is almost entirely about generation. We debate gigawatt targets, auction results, and capacity additions. We rarely ask whether we are using energy well in the first place. A genuinely holistic energy security framework has to go beyond “how much can we generate” and ask “how little do we actually need to generate.”

Megawatts vs Negawatts. A unit of energy saved is worth more than a unit of energy produced, because saved energy avoids not just generation cost but also transmission, transportation and distribution losses along the way. Yet our incentive structures, our headlines and our policy attention overwhelmingly reward the “Megawatt” — new capacity commissioned — and barely acknowledge the “Negawatt” — demand that was never created in the first place. Efficiency doesn’t get a ribbon-cutting ceremony, but it deserves serious policy weight.

Living with nature, not just extracting from it. Our energy strategy treats nature purely as a source to be exploited — sun, wind, water, coal — rather than as a rhythm to live within. Shifting working hours to track daylight, dressing for our own climate rather than inherited colonial dress codes, and designing buildings for local climate rather than imported architectural fashion are not fringe ideas — they are low-cost, high-impact ways to cut energy demand at the source. Right now, most of this sits in the “nice to have” bucket rather than being part of how we plan cities and workplaces.

Reimagining industrial townships. Much of our energy demand — and our transport-related energy waste — comes from workers commuting long distances between where they live and where they work. Integrated, energy-efficient industrial townships, designed so the workforce doesn’t need to travel far, would cut transport energy use meaningfully while also improving quality of life. This needs to be a deliberate part of industrial policy, not an incidental outcome of land availability.

The water-energy nexus. Water and energy are tightly linked in India, more than we usually admit. Pumping groundwater, moving water through canals, and growing water-intensive crops in water-stressed regions all consume large amounts of energy. Rainwater harvesting, smarter crop choices suited to local water availability, and better irrigation infrastructure would reduce the energy footprint of our water systems — an angle that rarely features in energy security discussions, but should.

Public transport as an energy strategy, not just an urban amenity. Every private vehicle trip that could have been a bus or metro trip is, whether we frame it that way or not, an energy security decision. Improving last-mile connectivity, raising the comfort and reliability of public transport, and providing real incentives to use it are among the highest-leverage ways to cut transport energy demand — one of the fastest-growing components of our energy consumption.

Why this matters

None of this is to dismiss the real progress India has made — the coal production milestone, the solar capacity build-out, and the narrowing, if still incomplete, of overall primary energy import dependence, which has fallen from about 47% to 42% between FY17 and FY25. But the numbers also make clear that generation-side progress alone will not get us to energy security. Our oil and gas production is structurally constrained. Our hydro potential remains underexploited. Promotion of pumped storage is a step in right direction. And even our solar success is currently being partly wasted for want of storage and grid flexibility.

A holistic energy security framework has to treat demand reduction, climate-appropriate living, and efficient urban and industrial design as being just as central to energy security as the next gigawatt of capacity. Until Negawatts get the same policy attention as Megawatts, India’s energy security will remain a story of running hard just to stay in place.

 

The next round of QE might break something

Last week, I found myself doing the most boring kind of research an investor can do: comparing grocery receipts, gathered from friends, across four cities. A basic basket — some fruit, dairy, a few staples, nothing fancy. The bill ran close to US$90 in San Francisco; US$32 in Singapore; about US$25 in Johor Bahru, just across the Malaysian border, and US$21 in New Delhi. Singapore is one of the most expensive cities on earth; but San Francisco was still ~40% pricier than that, and more than four times pricier than Johor and Delhi.

Those grocery receipts are a better macro indicator than most of what crosses my desk. Because if you strip away exchange rates and just ask what a US dollar, a Singapore dollar, a Ringgit or a Rupee can actually buy, the picture of the US economy that emerges is very different from the one in the headline GDP numbers. And once you sit with that picture for a while, a few other things I’ve been tracking — household debt, the shape of Treasury issuance, and the politics at the Federal Reserve — start to line up into a story that should worry anyone holding US assets on autopilot.

The GDP number is flattering itself

US GDP is measured in nominal dollars, and consumption is about 70% of that number. But nominal dollars in an economy where a basic grocery run costs three to four times what it costs across large parts of Asia are not the same thing as real purchasing power. See the US output through the purchasing-power-parity lens, which economists use to compare living standards across countries, and a meaningful chunk of America’s GDP lead evaporates. It isn’t that Americans are producing that much more; it’s just that everything they buy is priced that much higher.

This matters because Washington keeps pointing to a 3% year-on-year inflation print as evidence that the inflation problem is basically solved. It isn’t. 3% this year sits on top of a price level that has already doubled or tripled relative to large parts of the world over the last few decades. A “controlled” 3% on an already-inflated base is not the same as price stability — it’s a slower march up a mountain you’ve already climbed most of. When your starting altitude is this high, the next few percentage points matter a lot more than the official narrative suggests.

The debt picture looks better than it is

The standard rebuttal to any US-fragility argument is the debt chart: household debt as a share of GDP has fallen from roughly 98% at the 2008 peak to around 70% today (or from about 86% to 62% on the narrower Federal Reserve measure). In real, inflation-adjusted dollars, total household debt is still just below its 2009 high. On the surface, that looks like a system that deleveraged responsibly.

Look under the surface and the story changes. The composition of that debt has shifted away from mortgages — which are secured, long-dated, and mostly held by higher-income, higher-asset households — into student loans and auto loans, which are unsecured or attached to rapidly depreciating collateral, and which are concentrated in the bottom two-thirds of the income distribution. That’s not a healthier balance sheet; it’s the same debt burden redistributed onto households with the least capacity to absorb a shock.

 ​

​

 ​

​

 

And the shock is already showing up in the delinquency data. Student loan forbearance ended, and serious delinquencies jumped from under 1% to over 16% in a single year, with roughly one in ten dollars of student debt now 90-plus days past due. Credit card serious-delinquency flow is running near 7%, on balances carrying interest rates north of 22%. Auto loans have roughly doubled in dollar terms, stretched out to six- and seven-year terms on assets that lose value the moment they leave the factory, with negative equity now common. Put it together and the flow of debt going seriously delinquent across the household sector has nearly doubled year-on-year. That is not a slow-burn statistic — that is a system starting to crack at the load-bearing joints.

Why the next move is almost certainly more easing

Here is where the fiscal and monetary threads meet. The US Treasury has been quietly running its own version of quantitative easing by issuing an unusually large share of new debt — well over 80% — as bills with maturities under a year. That’s the fiscal equivalent of financing a house with an adjustable-rate mortgage instead of a thirty-year fixed: it looks cheap when short rates are low, and it becomes a rolling crisis the moment they aren’t. It works fine right up until it doesn’t.

Meanwhile, the buffer that used to absorb funding stress — excess bank reserves — has been drawn down. The last time reserves got this thin, in September 2019, the overnight repo market seized up and the Fed had to step in with emergency liquidity almost overnight. We are closer to a repeat of that than most portfolios are priced for.

Add the politics. Whoever ends up leading the Fed inherits an institution that is still majority hawkish, at a moment when the Treasury’s own borrowing structure and a fragile household sector are both quietly begging for lower rates and more liquidity. Any new chair with an inflation-hawk reputation has every incentive to hold the line as long as politically possible — partly on principle, partly to protect their own legacy — before the pressure becomes unavoidable. But the direction of travel is not really in question. Between an over-levered household sector, a Treasury that has to keep rolling short-dated debt, and a banking system running low on reserves, the destination is the same: more balance-sheet expansion, on a scale bigger than 2020 or 2008. The only real question is how much pain arrives in the gap between now and then.

The dangerous part is the gap, not the destination

This is the part that is genuinely hard to model, and it’s the part most investors are ignoring. Everyone can see the endpoint — easier money is coming, eventually, because the alternative is a funding crisis nobody in Washington wants to own. What’s much less certain is what happens in the interval between “everyone knows QE is coming” and “QE actually arrives.” That gap is where currency stress, bond-market air pockets, and credit events tend to live. History doesn’t move in a straight line from problem to solution; it moves through a chaotic middle where things break in ways nobody predicted, before policymakers are finally forced to act. Anyone who lived through 1997, 2000, or 2008 as an investor remembers that the crisis wasn’t the eventual policy response — it was the disorderly scramble that forced the response.

And if the policy response, when it comes, is aggressive — a real “super QE" rather than a modest top-up — the inflationary consequences could be severe. An economy already running elevated inflation on top of an inflated price base, hit with a large new wave of liquidity, does not necessarily settle back into 2–3% inflation. It can just as easily accelerate toward double digits, especially with a fiscal deficit that shows no sign of shrinking.

Why this time the social risk is not purely theoretical

Here is the part that doesn’t usually make it into investment commentary, but probably should. A generation of young Americans is coming of age with weak job prospects in traditionally high paying sectors, high living costs, record levels of unsecured consumer debt, and a wall of delinquencies already forming under their feet. Layer double-digit inflation on top of that combination, and you are not just looking at a market risk. You are looking at a social one.

Civil unrest as a tail risk for the US used to be the kind of thing you’d mention only half-seriously, a footnote in a scenario-planning deck. I don’t think it’s purely theoretical anymore. It doesn’t need to become the base case to matter — even a modest, non-trivial probability of serious social instability is the kind of risk that reprices currencies, sovereign spreads, and equity risk premia long before it shows up on the evening news. Markets are usually terrible at pricing tail risks like this until the tail starts wagging the dog.

What this means for how I’m positioning

None of this is a call to panic or to abandon US assets wholesale — the United States remains the deepest, most liquid capital market in the world, and that doesn’t change overnight. But it is a case for taking seriously a scenario the consensus is currently pricing at close to zero: a period of genuine monetary, fiscal, and social stress in the US, sitting between now and the eventual “everything gets easier” resolution. That argues for holding real assets and inflation hedges; for not over-extending duration into US paper on the assumption that rates only fall from here in an orderly way; and for treating US-dollar cash as a tool rather than a safe haven by default.

The lesson from every cycle I’ve watched — 1997, 2000, 2008 — is the same: the eventual policy response is rarely the dangerous part. The dangerous part is the disorderly stretch of road that forces policymakers into that response in the first place. We may be entering that stretch of road now, and it’s worth positioning for the possibility that it gets bumpier before it gets smoother.

 

The Balloon in the Dark Room

Yesterday’s post drew a wider response than I expected, and I am grateful for it. Most readers found the argument thought-provoking and timely; a smaller number found it unnecessarily alarming and short on substance. Both reactions are fair, and I want to address the second group directly rather than let the disagreement sit unanswered.

When the numbers stop explaining themselves

Every investing career eventually runs into a stretch where the spreadsheet stops being a reliable guide. Valuation models still produce an output, but the confidence behind the output erodes, because the inputs themselves — growth rates, discount rates, terminal assumptions — are being asked to do work they were never designed for. In those stretches, experienced investors tend to fall back on something closer to pattern recognition than arithmetic: a sense, built from having sat through a few cycles, that the character of the market has shifted even before the numbers confirm it.

My sense is that we are in exactly that kind of stretch. It is not a single data point driving that view; it is the accumulation of several stress signals appearing at the same time, each individually explainable, but collectively unusual.

The signals worth naming

Fiscal sustainability. In several major economies, government deficits and debt-servicing costs have reached levels that would once have triggered a sharp market response. Instead, they have been absorbed with surprising calm, largely because interest rate cycles and demand for sovereign debt have so far cooperated. That cooperation is not guaranteed to continue indefinitely.

Sticky inflation. Headline inflation has come off its post-pandemic peaks, but the underlying components — services, wages, shelter — have proven far stickier than central banks initially projected. This matters because it narrows the room policymakers have to cut rates aggressively if growth weakens, which is precisely the tool that has rescued markets in past downturns.

Stagnating growth. Several developed economies are showing growth rates that are positive on paper but weak beneath the surface — propped up by government spending, a handful of mega-cap companies, or one-off factors, rather than broad-based private investment.

Stretched growth valuations. A narrow set of companies, heavily concentrated in AI-linked infrastructure, is being priced as though its current growth trajectory extends cleanly for a decade or more. Multiples of this kind have historically required near-perfect execution to be justified, and near-perfect execution is a high bar over a long horizon.

Rising credit stress. Pockets of the credit market — leveraged loans, private credit, and select corporate sectors — are showing early signs of strain: widening spreads, downgrades, and refinancing difficulty. Credit stress is usually a leading indicator rather than a lagging one; it tends to show up before equity markets acknowledge a problem.

Geopolitics and trade. Tariff uncertainty, shifting alliances, and logistics disruption are adding a layer of unpredictability to corporate planning that did not exist to the same degree a decade ago. Supply chains built for efficiency are being rebuilt for resilience, and that transition is expensive and slow.

Volatility across unrelated assets. Semiconductor equities, silver, bitcoin, and the SpaceX IPO have each delivered sharp, painful drawdowns in recent months. What is notable is not any single move, but that the losses have appeared across asset classes that are not normally correlated — a sign that liquidity and risk appetite, not asset-specific fundamentals, may be the common thread.

AI’s unresolved business model. Artificial intelligence has genuinely disrupted several legacy industries, and that part of the story is real. What remains unproven is whether the AI buildout itself — the capital expenditure, the compute, the infrastructure — will generate a return that justifies its cost on any reasonable timeline. Disruption of others’ business models is not the same as having a durable one of your own.

The balloon in the Dark Room

None of these signals, on its own, is a crisis. That is exactly what makes the moment harder to read, not easier. It feels like a group of persons standing in a dark room holding pins in their hands, knowing there is a balloon somewhere above, without knowing which pin — or whose hand — will find it first. The probability that no pin ever finds the balloon is not zero. But it is not the base case either, and treating it as the base case is where I think the complacent reading of this market goes wrong.

Why this could be a combination, not a repeat

Markets rarely repeat a crisis exactly; they tend to borrow ingredients from more than one. My working view is that if a serious correction unfolds from here, it is more likely to resemble a blend of three earlier episodes than a clean replay of any single one.

2008 and 2000, combined. The 2008 crisis was fundamentally a credit crisis — leverage built on assets whose value was assumed rather than tested. The 2000 crisis was a story of capital pouring into business models that had not yet proven they could generate durable cash flow. The current AI investment cycle carries elements of both: enormous capital expenditure, increasingly financed with debt and vendor-financing arrangements, directed at business models that remain commercially unproven at scale.

1997 and 1998, revisited. The Asian financial crisis and the collapse of Long-Term Capital Management were both, at their core, stories of a strengthening US dollar exposing leverage and mispricing elsewhere in the system. A materially stronger dollar today would put similar pressure on precious metals and crypto assets, both of which have attracted significant speculative positioning over the past two years.

1987, echoed. Black Monday was less about a single fundamental trigger and more about mechanics: rising rates, unresolved trade and currency disputes, stretched valuations, and — critically — automated trading systems that amplified a decline rather than cushioning it. The scale and speed of algorithmic execution in markets today is far greater than it was in 1987, which means the mechanical amplification risk, if anything, is larger, not smaller.

A downturn that borrows credit stress from 2008, unproven business models from 2000, dollar-driven deleveraging from 1997, and mechanical amplification from 1987 would not look identical to any one of those events. It could plausibly be sharper and more disorienting than any of them individually, precisely because it would be arriving from several directions at once rather than one.

What the twelve months before every crash have in common

It is worth revisiting how each of these episodes felt in the twelve months before they began, because the emotional texture tends to rhyme even when the underlying causes differ. In 1996, 1999, and 2007, the dominant mood was not fear — it was irritation with anyone who raised the possibility of a top. Alan Greenspan’s irrational exuberance remark in December 1996 was widely dismissed at the time, and the market ran for several more years before the reckoning arrived. The same pattern held in 1999, and again in 2007, when concerns about housing and credit quality were treated as the fringe view rather than the mainstream one.

I do not think 2026 sounds meaningfully different from those years. The specific worry differs each cycle, but the response to the worry — dismissal, followed by irritation, followed eventually by recognition — has been remarkably consistent. That consistency is itself part of the argument, not a footnote to it.

A closing note

None of this is a prediction with a date attached, and it should not be read as one. It is a personal, opinion-based on reading of conditions that I believe warrant more caution than the current mood in markets reflects. Markets can, and often do, remain irrational for longer than any individual view accounts for, and I could simply be wrong. Regardless, I would like to be cautious and brace myself for the “reset”. Also, I completely respect the disagreement.