If The S&P Always Goes Up, Why Not Buy It Leveraged?

TL;DR

• Leveraged ETFs deliver 2x or 3x the index's return per day, not per year — and that difference is the entire story.
• Daily resetting means volatility itself eats your money: an index that goes +10% then −9.1% is flat; the 3x version is down 5.5%. Sideways chop grinds leverage to dust.
• Deep crashes are near-fatal by arithmetic: 3x a −34% month (2020) is close to −100%, and −90% needs a +900% recovery. A simulated 3x Nasdaq fund through 2000–2002 loses ~99.9% and doesn't recover for decades.
• "The S&P always goes up over time" is true. Leveraged funds don't live "over time" — they live day to day, charge ~1% fees plus borrowing costs, and are built (and marketed) as trading tools, not holdings.

The Seductive Logic

The pitch assembles itself: the S&P 500 has averaged ~10% a year for a century and has recovered from every crash in history. If a thing reliably goes up, borrowing to own more of it is just… efficiency? A 3x fund like UPRO should turn 10% a year into 30%, and the crashes are survivable because the index always comes back. Every investor builds this argument in the shower at some point. Some decades — 2010–2021, for instance — even reward it spectacularly, which is what makes it dangerous rather than merely wrong.

The flaw isn't in the premise about the S&P. It's in the word "always goes up over time." Leveraged ETFs do not experience time the way you do.

The Daily Reset

A "3x S&P 500" fund promises three times the index's return each individual day. Every morning it rebalances its borrowing so today's move is tripled from today's starting point, with yesterday forgotten. Over any period longer than a day, your return is those daily triples compounded — which is not 3x the index's period return. Sometimes it's more; often, and especially in rough seas, it's much less.

Watch the arithmetic do it. The index goes +10% one day, −9.09% the next: it's back to exactly 100. Flat. The 3x fund goes +30%, then −27.27%: 100 → 130 → 94.5. The index round-tripped; you lost 5.5%. Nothing malfunctioned — compounding sequences of tripled ups and downs is simply lossy, and the loss scales with how much the path wiggles. Hence the name: volatility decay. The market chops sideways in a ±1.5% range for three months and the unleveraged holder is flat while the 3x holder bleeds continuously, confused, having been right about the market and wrong about the vehicle.

So the honest formulation is: leveraged ETFs multiply daily direction, but over longer horizons you're short volatility whether you knew it or not. Smooth grind upward: leverage over-delivers (2023–24 style, TQQQ up ~200% in a year — the ads write themselves). Volatile anything else: it under-delivers, sometimes catastrophically.

The Crash Math

Ordinary losses need outsized recoveries — a 50% loss needs +100% back. Leverage turbocharges the left side of that ledger:

• 2020, February–March: S&P −34% in a month. UPRO (3x): about −75%. The index needed +52% to recover; the fund needed +300%. The index reclaimed its high in six months; leveraged holders waited much longer — and that was the lucky, V-shaped crash.
• 2022, a mere bad year: S&P −19%. UPRO roughly −57%, TQQQ (3x Nasdaq) about −79% — a full-scale wipeout extracted from a garden-variety bear market, courtesy of decay stacked on decline.
• 2000–2002, the nightmare case: the Nasdaq fell ~78% over two grinding years. Simulate a 3x Nasdaq fund through it (TQQQ didn't exist yet — telling, we'll get to it) and you lose ~99.9%. From −99.9%, recovery requires +100,000%; a simulated dot-com-peak TQQQ buyer is underwater for decades. "The index always comes back" is true and useless — you come back only if there's something left to compound. At −99.9% there isn't.

And a slow-motion detail people miss: these funds triple daily moves, so a 1987-style single day (−20%) means −60% before lunch — it's why some 3x funds were designed with that scenario in the fine print. The S&P's recoveries happen on the index's schedule, in years; leverage's losses happen on compounding's schedule, immediately.

The Quieter Costs

Even in kind markets, the deck rents against you. Expense ratios run ~0.9% (UPRO 0.91%, TQQQ ~0.86% — 30x a plain index fund), and the fund pays financing costs on the borrowed 200% at roughly short-term rates — when cash pays 4–5%, that's ~8–10% a year of drag before decay, versus the 0.03% index fund. The issuers themselves label these products as short-term trading tools and publish the daily-reset warnings; regulators make brokers flash disclosures for a reason. TQQQ launched in 2010 — the start of history's friendliest decade for exactly this product. Its marketing chart begins, conveniently, after the event that would have killed it.

The Honest Version Of The Answer

"Why not leveraged S&P?" — because you'd be converting a bet you can't lose over 30 years into a sequence of 7,500 daily bets you can absolutely lose, while paying ~10x-to-300x more in costs for the conversion. The unleveraged index's superpower is that its worst historical outcome, held long enough, was "wait a while." Leverage forfeits exactly that property — the one that made the S&P a sure thing in the first place.

(For completeness: academics have studied modest leverage — up to ~2x — early in life as "lifecycle investing," and it's not insane on paper, with borrowing costs, discipline, and decades of runway. In practice, the people asking this question after seeing TQQQ's chart are not executing a Yale paper; they're chasing the fast version of a slow thing, and the market bills for that reliably.)

The unlevered answer remains undefeated: a broad index fund, time, and the leverage everyone forgets they already have — future paychecks, invested monthly.