Leverage is the use of buying power beyond the assets a trader or investor currently controls, allowing investors to bet larger with the same initial capital base. There are different ways to obtain leverage – the main three in the crypto world being lending desks, derivatives exchanges, and lending protocols. While this article focuses mostly on leverage obtained from derivatives exchanges, the high-level concepts are the same everywhere.
In the context of trading specifically, leverage is the ratio of your position size and your account value (also called “margin”).
For example, assume BTC is at $50k, and you buy 1 BTC worth of a futures contract.
- If your account has $10k of margin, your leverage is $50k / $10k = 5x.
- If your account has $5k of margin, your leverage would be twice that, or 10x.
With leverage comes the opportunity to make outsized gains (and losses) with a small capital base.
What are the dynamics of leverage?
It is important to understand how the use of leverage changes your positions, and the way you must manage them. Since leverage = position size/margin, your leverage will change if either the numerator or denominator changes.
Assume again a 1 BTC long position when BTC is at $50k, with $10k of margin. Hence your leverage is 5x.
If BTC moves to $60k, what happens?
For one, the value of your position goes to $60k, creating positive PNL of ($60k – $50k) = $10k. The PNL adds to your margin, which is now $10k + $10k = $20k, lowering leverage to 3x.
If BTC instead dropped to $45k, your leverage would be $40k / ($10k + $45k – $50k) = 8x.
So you can see, as the market moves in favor of your position, your leverage decreases. But if it moves against your position, leverage increases. This makes leverage quite dangerous. If your leverage gets too high, you run the risk of liquidation.
Leverage and PNL
Another important thing to remember is that the amount of leverage used says nothing about a trade’s profit potential. PNL is entirely the result of position size.
A 1 BTC long position will always win or lose the same amount, whether the position is backed by 10% or 100% margin. All that changes is the liquidation threshold, i.e. how much you can lose on the downside.
Whenever you want to make a new directional bet, do not think about leverage. Instead, you want to think about how to size your bet in the context of your overall portfolio using a framework such as the Kelly Criterion.
Thinking about the position size always comes first, only then followed by how to put on the trade in the most efficient way.
Managing the risks of leverage
Since leverage involves using capital that you do not have a claim to, there is a risk that comes along with using it – liquidation. In the traditional markets, it is possible (with leverage) to lose more money than you put into the markets. If you borrow 40,000 with 10,000, and happen to bet on a stock that goes to zero — tough luck. You’re legally on the hook for 40,000.
In the crypto world, there is little legal recourse for exchanges against anonymous customers, so this problem is solved via liquidation. In exchange for not being able to lose more than you put in – you incur the risk of liquidation.
First, you have to understand the concept of initial margin, and maintenance margin. Initial margin is the capital required to initiate a position. For example, if the initial margin for a position is 1%, you need to front 1% of the full 100% position. This means you can take out 100 to 1 leverage. If the initial margin was 50%, you would only be able to take out 2:1 leverage.
Once the position is established, you are subject to maintenance margin. Maintenance margin is the amount of capital relative to position size you must have in order to keep the position open. If your ratio of account equity to position size goes below the maintenance margin, you get liquidated.
Before liquidation, traders can manually reduce their position (to lower the numerator) or add more margin to their account (to raise the denominator), both of which decrease leverage.
You should really try not to get liquidated because it equates to donating money to the exchange. Below the maintenance margin, most exchanges will internalize your position (buy it from you) and then try to close it against the market.
The maintenance margin tends to be a fair bit higher than the true bankruptcy price to give the exchange some padding. Any excess goes to the insurance fund, which is basically profit to the exchange with extra steps.
To prevent that, it is always better to put a stop order somewhere above the liquidation price.
Finally, you have to choose between using isolated margin vs cross margin mode for your account. With isolated margin, a specific amount of margin is allocated to each trading pair. In cross margin, all trading pairs can access the same unified margin pool.
Which choice is right is a matter of tradeoffs. In isolated margin, the risk for each position is siloed and hence cannot accidentally liquidate other positions – thereby preventing a common beginner’s mistake. This is because the exchange guarantees your stop level, which is strictly safer than setting a stop loss on cross margin.
However, in cross, you can have offsetting positions, such as long BTC future / short ETH future. So cross margin is riskier, but also more capital-efficient.
Why use leverage?
So with all the risks, why use leverage?
Asymmetric directional bets
Given that leverage increases both the upside and downside potential of a directional trade, you only want to use it when expecting a big upside, but little downside.
One such example would be when there is new positive information (e.g. news) that hasn’t permeated the market yet, but you are certain of. You may believe that there is a short window of time to take advantage of this and that the outcome will either be positive or neutral – but very unlikely to be negative. Your confidence in the timing of the trade being short-term also reduces the risk that standard noise will take you out of your position.
In that situation, you could consider taking out a leveraged bet because you believe there is significant asymmetry in the bet.
Delta neutral strategies
Leverage is extremely attractive when running delta-neutral strategies. These are strategies, such as arbitraging different maturities of the same product, or performing price arbitrage across exchanges, that have little to no directional exposure.
Depending on the trade, the borrower is less likely to get liquidated and can use leverage to juice more returns out of existing trades at little additional risk.
That said, even strategies that are risk-neutral on a portfolio level are often not so on a per exchange level.
Reduce counterparty-risk, increase capital efficiency
Often leverage can be used to increase the efficiency of trading in your portfolio. For example, you may choose to take out leverage on Bitcoin to recreate exposure, sell some of your spot Bitcoin to cash, and generate a yield higher than the cost of leverage with the cash generation from the sale.
To put numbers to this, imagine you own 1 Bitcoin (worth $50k) and want to keep 1 Bitcoin of exposure. But you notice that there are yield opportunities that will pay you 80% a year if you deposit cash. You can sell 0.5 Bitcoin into 25,000 USDT, and 2x lever on a futures product with the remaining 0.5 to recreate your exposure. With the 25,000 USDT, you deposit into a yield farm that generates 80% a year, covering your cost of leverage. Of course, you should be mindful of getting liquidated.
Comparing different types of leverage
Traders can get leverage in four different ways.
- Spot margin
- Perpetual swaps
They differ primarily in two dimensions: funding cost, and stop-out risk.
When using spot margin, the trader needs to borrow physical coins or USD to increase their buying power, so there must be a lender. This role can be either filled by other customers in a p2p lending market or by the exchange itself.
This is in contrast to the other three products, where no actual lending and borrowing takes place. Swaps, futures, and options are financial contracts on the outcome of a bet on the underlying asset. A new contract is created from thin air whenever a buyer and a seller are matched by the exchange.
As a result, there is no need to collateralize the entire position. There simply need to be rules on how the contract’s PNL is calculated, and then counter parties only exchange the PNL that they actually owe. This higher capital efficiency is a major benefit of derivatives vs. spot.
Decentralized lending protocols such as Aave or Compound also allow users to take out spot margin against collateral. Often the collateral requirements are higher for decentralized platforms, but the borrowing cost can sometimes be cheaper.
In perpetual swaps, longs and shorts exchange funding payments based on two factors. First, the distance of the mark price from the index price. And second, a virtual interest rate. A long is functionally borrowing USD from the short side, while the short side is borrowing BTC from the long side. USD is typically more expensive to borrow than BTC, that’s why longs have to pay shorts a small interest rate when the market is flat.
In futures, the funding cost is in the difference of the futures price from the index, called basis. This difference is reported in annualized form, with the shorter-dated futures often priced higher than longer-dated futures (in the crypto markets). For example, at the time of writing the Deribit September Future is trading at ~$55k, compared to the current spot price of ~$48k. That implies you pay an annualized rate of 24% a year to hold the future. The benefit of buying the future is that you lock in a rate, where-as the perpetual has a variable rate.
With options, you can control your leverage by selecting options with a different strike price. Each option contract on Deribit represents one notional Bitcoin, but the amount of “exposure” you are taking to price relies entirely on the “delta” of the option.
The further out of the money the option currently is, the cheaper it will be priced, and hence the more notional exposure you can buy with the same collateral. To calculate the cheapest way of getting leverage though, one must divide the delta of the option by the premium of the option. Out-of-the-money options tend to offer more leverage than at-the-money options. To calculate your overall leverage using options, take the total delta of the position plus your unused Bitcoin, and divide by the overall equity of the account. For example, if you have 1 Bitcoin in your Deribit account, and you buy 0.5BTC worth of options that sum up to 2 delta, your leverage would be (2 + .5) / 1 or 2.5x leverage.
Stop-out risk is the risk involved with closing your position.
Spot margin, perps, and futures carry larger risks than options in that context because to close them, you need to get a market order filled. However, the price at where you can get out is often hard to predict in advance, especially when the market crashes or spikes.
During such moves, liquidity often evaporates from the orderbook as MMs widen their bid-ask spread, and the stop orders and liquidations of other participants exacerbate the market move in a way that gets even more positions liquidated.
As a result, traders can get filled way far beyond their original stop, and sometimes not at all.
Longing call and put options on the other hand has a clearly defined risk, as the trader can never lose more than the premium they paid for it in the first place.
Leverage is a powerful tool that allows betting larger with the same initial capital base. Because leverage also magnifies downside risk, it should be reserved for trades with low risk – whether they be very asymmetrical directional bets or delta-neutral strategies.
We showed that different ways to create leverage have unique tradeoffs in terms of stop-out risk and funding cost. Before using any of them, you should be familiar with the risk.