Imagine a contract you can hold forever, with no expiry date — and a rule that says one side must pay the other a small fee every few hours, around the clock. If you arrange to be on the 

 side of that fee while taking no view at all on whether the market rises or falls, you get paid — again and again — simply for holding a position.

That payment is real, and it powers an entire category of market-neutral crypto strategies. It's called the 

A normal futures contract has an expiry date. On that date, the future and the spot price must meet — that forced meeting is what makes a basis trade work.

 (everyone calls it a "perp"). You can hold it forever. It's the single most traded instrument in crypto.

But this creates a problem. A normal future is 

 toward spot by its expiry. A perp has no expiry, so what stops its price from floating away from the real spot price of Bitcoin? Nothing — unless you invent a force to hold it in place.

 a perp is a futures contract that never expires. Because it never expires, it needs a special mechanism to keep its price glued to spot.

The funding rate is a small payment made directly between traders — 

The direction depends on which side is too eager:

This is a clever self-correcting system. If everyone crowds the long side and pushes the perp above spot, longs get 

 (they receive it) — which nudges traders back toward balance and drags the perp back to spot. The fee is the leash.

 every 8 hours, whichever side is crowded pays the other side a small fee. That fee is what keeps the never-expiring perp from drifting away from the real price.

The size is small per payment — a common "neutral" level is about 

 — but it repeats three times a day, every day. Annualise it and 0.01% × 3 × 365 ≈ 

 And in a hot bull market, when longs are desperate for leverage, funding can run many times higher.

3. The carry trade: be the side that gets paid

, shorts get paid. So if you could be short 

 actually betting that the price falls, you'd collect that stream for free.

The two cancel. If BTC doubles or halves, your spot gain is your perp loss, or vice versa — you don't care which way the market goes. You are 

 every 8 hours. So you sit there, fully hedged, collecting funding three times a day. That's funding-rate carry.

 buy the coin and short the same amount of perp. The price moves cancel out, but you still collect the funding fee — a market-neutral yield for holding still.

Notice the shape of this: long the asset, short the derivative against it. It's the classic buy-and-hedge structure — except instead of waiting for a fixed gap to close at some expiry date, you're harvesting a recurring payment with no end date at all.

Each funding payment: 0.01% of $60,000 = 

, on about $60,000 of capital. That's the ~11% annualised yield, and you carried 

Now picture a frothy bull market where funding spikes to 

. The same position now earns 0.05% × 3 × 365 ≈ 

 is why delta-neutral crypto desks and "market-neutral yield" funds exist: they're industrialising this exact trade across many coins.

 at calm funding you earn around 11% a year; in a leverage-hungry bull market the same hedged position can earn far more — all without betting on price.

5. The reverse case (and why it's harder)

 — when shorts pay longs? To harvest that, you'd flip the trade: 

 To sell a coin you don't own, you'd have to borrow it first — which is costly and often simply unavailable. That makes the negative-funding harvest the awkward, less-accessible direction. Positive-funding carry (long spot, short perp) is the one most people can actually run.

 funding means shorting the coin itself, which is hard and pricey — so the easy, common version is the positive-funding one.

6. The catches — why "free yield" isn't free

This trade is genuinely elegant, but it has teeth.

 11%. Funding changes every 8 hours and can 

 — at which point your lovely income becomes an 

 paying. The yield you signed up for can quietly turn into a slow bleed. You can exit, but it's a position you must babysit, not a fixed deposit.

Liquidation risk on the short leg — the killer.

 Here's the trap that catches beginners. Your position is delta-neutral 

, but the two legs live in different places. If BTC rips upward, your short perp shows a big loss 

 — even though your spot is up by the same amount, that gain may be sitting elsewhere and not counted as collateral. If you haven't kept a margin buffer, the exchange 

 at the worst moment, and suddenly you're left holding only the spot leg — fully exposed, the hedge gone. A "market-neutral" trade can absolutely blow up this way. Managing collateral is the whole job.

 Your money sits on an exchange. If it freezes withdrawals or collapses (Mt. Gox, FTX), the yield is the least of your worries. Spreading across venues helps but adds complexity.

 Because you enter once and hold rather than trading constantly, the 1% TDS on crypto trades only hits your entry and exit — so it isn't multiplied into oblivion. But it's still there, gains are taxed at a flat 

 with no offsetting of losses, and the tax treatment of the funding receipts themselves is genuinely murky. On a true ~11% gross yield, Indian taxes can take a large bite out of what's left.

 Two legs to open and two to close. On a short holding period, those fills can outweigh several days of funding. The carry needs 

 the yield can flip negative, your hedge can get liquidated if you don't watch your margin, your money sits on a fragile exchange, and Indian taxes take a real cut. Elegant — but not a sleep-easy trade.

7. For the coders: a funding-carry scanner

The maths is small. Annualise each coin's funding, subtract what it costs to run the trade over your horizon, and rank what's left. The honesty, as always, is in the subtraction.

def annualise_funding(rate_per_8h_pct, periods_per_day=3):
    """Turn an 8-hourly funding rate (%) into an annual yield (%)."""
    return rate_per_8h_pct * periods_per_day * 365

def net_carry(rate_per_8h_pct, fee_per_leg_pct=0.05, holding_days=30):
    """
    Annualised net carry of long-spot / short-perp, after fees.
    Opening + closing = 4 fills (2 legs in, 2 legs out).
    A one-off cost, spread (annualised) over the holding period.
    """
    gross = annualise_funding(rate_per_8h_pct)
    roundtrip_cost = fee_per_leg_pct * 4                 # one-off, % of notional
    cost_annualised = roundtrip_cost * (365 / holding_days)
    return round(gross - cost_annualised, 2)

# a small watchlist of 8h funding rates (%)
funding = {"BTC": 0.010, "ETH": 0.015, "SOL": 0.040, "DOGE": -0.008}

print(f"{'COIN':5} {'gross %/yr':>11} {'net %/yr (30d hold)':>22}")
for coin, rate in funding.items():
    g = round(annualise_funding(rate), 2)
    n = net_carry(rate, holding_days=30)
    print(f"{coin:5} {g:>11} {n:>22}")


COIN   gross %/yr    net %/yr (30d hold)
BTC         10.95                   8.52
ETH         16.43                  13.99
SOL          43.8                  41.37
DOGE        -8.76                 -11.19


The lesson lives in the last two columns. SOL's hot funding throws off a fat carry; DOGE has 

 funding, so the long-spot/short-perp trade 

 (you'd want the hard-to-run reverse). And notice the gap between gross and net shrinks the longer you hold — because the one-off fees get spread thinner. A real version would also pull live funding 

 the next predicted funding, check that both legs are liquid enough for your size, and — most importantly — model a margin buffer so the short can't get liquidated in a spike.

 annualise the funding, subtract the fees over your holding time, drop anything negative. The big remaining job the code must handle is keeping the short alive through price spikes.

8. The hidden bonus: funding is a sentiment gauge

Even if you never run the trade, the funding rate is one of the best free sentiment meters in crypto. 

High positive funding means longs are crowded and paying through the nose to stay long

 — a classic sign of greed, and often the setup right before a violent "long squeeze" flushes everyone out. Deeply negative funding means the opposite: shorts are crowded and paying to stay short, which can precede a short squeeze. Reading funding tells you where the leverage is piled up — and crowded leverage is where the fireworks happen. That's useful market intelligence whether or not you ever collect a single payment.

Crypto perpetual futures never expire, so exchanges use a funding rate — a small fee paid every 8 hours between longs and shorts — to keep the perp glued to spot. When funding is positive, shorts get paid, so you buy the coin and short an equal amount of perp: the price moves cancel (market-neutral) but you keep collecting the funding, often around 11% a year and far more in a hot bull market. It's a buy-and-hedge position with no expiry — you simply collect the recurring funding instead of waiting for a gap to close. The catches are real: funding can flip and start costing you, your hedge can be liquidated if you don't manage margin, your capital sits on a fragile exchange, and Indian taxes take a cut. And even if you never trade it, funding is a superb gauge of where leverage — and danger — is building up.

Educational only, not investment advice. Leveraged and perpetual-futures positions can be liquidated and carry real risk of loss, including total loss; consult a registered advisor before trading.