How to Compare Rates on Crypto-Backed Loans

The short answer: you cannot compare crypto-backed loans on their advertised rates, because no two lenders put the same costs inside that number. To compare them honestly, add the interest rate to every fee the lender discloses, convert any fee charged on collateral into a percentage of the amount you actually borrow, and hold the loan size, loan-to-value and term constant across every lender. That figure — the illustrative annualized borrowing cost — is the only one that compares like for like. This article explains how to calculate it, where the underlying numbers come from, and what to weigh besides price. The live comparison below is refreshed weekly.

Why the advertised rate is rarely what you pay

A crypto-backed loan is a secured loan: you pledge Bitcoin or another digital asset, draw dollars against it, and keep the asset unless its value falls far enough to trigger liquidation. Because the loan is collateralized and there is usually no credit check, lenders compete loudly on the interest rate.

The difficulty is that “rate” is not standardized in this market the way APR is in mortgage lending. Four kinds of cost routinely sit outside the advertised number.

Origination and draw fees

A one-time charge on the amount you draw. On a twelve-month loan, an origination fee translates almost point-for-point into annualized cost: a two-point fee adds two points to what the loan costs you over a year, which is frequently larger than the difference between two lenders’ headline rates. Two details are worth checking. First, whether the fee applies to every draw or only the first — on a revolving line, a per-draw fee changes the economics of using it. Second, whether it is deducted from the proceeds, in which case you receive less than you borrowed while paying interest on the full amount.

Fees charged on collateral rather than on the loan

This is the one that most often escapes notice, and it is structural rather than a matter of degree. A fee assessed on your collateral is not comparable to the same percentage assessed on your loan. If you have posted twice your loan in collateral to sit at 50% loan-to-value, a fee on collateral costs you double its stated rate as a share of the money you actually received. At 25% loan-to-value it costs four times. The effect runs the opposite way to intuition: the more conservatively you borrow, the more these fees cost you in relative terms. We return to this below, because it applies to Abra’s own pricing.

Conversion, wrapping and custody costs

Several lenders do not hold your Bitcoin as Bitcoin. Some convert it to a wrapped representation to use it on a particular chain; some wrap it for deposit into lending protocols. Each round trip usually carries a fee. More importantly, each introduces a dependency on infrastructure beyond the lender itself, which is a risk question as well as a cost question.

Tiering you may not qualify for

Advertised rates are commonly the best rate in a tiered structure, and the tiers are what determine whether that rate is available to you. Three tiering models are in use, and they are worth telling apart because only one of them is within your control at the point of borrowing:

• Tiered by loan size. The best rate requires a large loan. Rate cards are usually published; find the bracket your loan actually falls into rather than reading the top line.
• Tiered by loan-to-value. The best rate requires a conservative position, which may be a trade-off you were happy to make anyway.
• Tiered by token holding. The best rate requires holding a quantity of the lender’s own token. This makes the advertised floor unavailable to anyone unwilling to take on that separate, unrelated exposure, and it is worth treating an advertised rate of this kind as a different product from one you can access on day one.

A lender may also quote a “from” rate that is the lowest across several collateral assets. If you intend to pledge Bitcoin, confirm the rate applies to Bitcoin specifically.

How to calculate annualized borrowing cost

Once you have the components, the arithmetic is simple. Annualized borrowing cost, as a percentage, is the sum of:

• the interest rate
• plus the origination fee, as a percentage of the loan
• plus annual fees charged on the loan
• plus annual fees charged on collateral, divided by loan-to-value
• plus one-time fees charged on collateral, divided by loan-to-value

Dividing by loan-to-value is what converts a collateral-based fee to a loan-based one. At 50% loan-to-value you divide by 0.5, which doubles it.

A worked example

The figures below are round numbers chosen to show the mechanics. They are not any lender’s actual pricing — for that, see the live comparison. Take a hypothetical lender advertising a 5.00% rate, with a 1.00% origination fee on the loan, a 0.20% annual fee on collateral, and a 0.30% wrapping fee also on collateral. On a $200,000 loan at 50% loan-to-value, meaning $400,000 of collateral, held for twelve months:

Illustrative only. Not any lender’s pricing.

A 5.00% rate costs 7.00% to carry. Two of the four line items landed at double their stated percentage purely because they are charged on collateral. Nothing about that two-point gap is visible on a pricing page.

Checking your own arithmetic

Where a lender publishes both an interest rate and an APR, you have a free check on your method: APR is a regulated disclosure that already incorporates fees, so your calculated figure should land close to it. If it does not, you have either missed a fee or misread its basis. Bear in mind that a published APR is calculated on a particular loan size, often a small one, so expect a modest difference rather than an exact match. Lenders who publish no APR are not necessarily more expensive — they are simply less directly comparable, which is the whole reason for doing this calculation yourself.

Borrowing costs compared

The table below holds the scenario constant — a $250,000 loan at 50% loan-to-value held for twelve months — and shows each disclosed cost alongside the terms that govern the loan. Figures are compiled from each lender’s own published terms and refreshed on the cadence stated with the comparison. Cells reading “not published” are exactly that: we have left them empty rather than estimate them, and the section on sources below explains which fields are missing and why.

Rates and terms change frequently and without notice. Confirm current terms directly with any lender before borrowing.

Why your loan-to-value changes which lender is cheapest

Because some lenders charge fees on the loan and others on collateral, the ranking is not fixed across the loan-to-value range. A lender whose fees are charged on the loan has a cost that stays flat as your loan-to-value changes. A lender whose fees are charged on collateral has a cost that rises as your loan-to-value falls, because the same loan is now supported by more collateral.

The practical consequence: there can be a crossover point below which a lender with a higher headline rate becomes the cheaper option overall. This matters most to the borrower who deliberately over-collateralizes to keep a wide buffer above the liquidation threshold — exactly the borrower who is being careful, and exactly the borrower most likely to assume that being careful is free.

Abra’s own fee structure is collateral-based in part, so this effect works against Abra at low loan-to-value. Work the arithmetic above at your own loan-to-value rather than reading the headline figure.

What to weigh besides the rate

Cost is one input. Several structural characteristics matter as much or more, and unlike rates they rarely change — they are properties of how a lender is built, so they are worth deciding on before you compare prices at all.

Fixed or variable

A variable rate can rise after you have borrowed; a fixed rate cannot. This is a genuine fork rather than a gradient, and no current rate advantage substitutes for certainty if certainty is what you need. Abra’s rate is variable, derived from an underlying lending market. If you need to know your cost for the next twelve months with confidence, a fixed-rate lender is the better fit regardless of where the numbers sit this week.

Maximum loan-to-value

This determines how much cash a given amount of collateral can raise, and the range across lenders is wide. If your binding constraint is how much you can borrow rather than what it costs, start by filtering on maximum loan-to-value and only then compare price among the lenders that clear your threshold.

Collateral accepted

Some lenders take a single asset; others take several. This is a qualifying question, not a preference: if a lender does not accept what you hold, its rate is irrelevant to you. Abra accepts a narrower set than some of the alternatives in the table.

Custody and protocol risk

Ask where your collateral actually goes. Models in use include qualified custodians, decentralized MPC custody, and deployment into DeFi lending protocols. Collateral deployed to a protocol carries smart-contract and protocol risk on top of the ordinary risks of a collateralized loan. Abra wraps Bitcoin collateral for deposit into DeFi lending pools, so a borrower who regards protocol risk as inadequately compensated by a lower rate should weigh that against us, and we do not think that is an unreasonable position to hold.

Liquidation mechanics

Three separate things are worth establishing, and they are frequently disclosed at different levels of detail: the loan-to-value at which you receive a margin call, the loan-to-value at which collateral is actually sold, and what it costs when that happens. Trade spreads, liquidation penalties and optional liquidation-protection products only ever apply on your worst day, which is precisely why they repay reading in advance. A high maximum loan-to-value paired with an undisclosed liquidation threshold deserves a direct question.

Term structure and repayment

Open-term loans have no maturity date and generally no prepayment penalty. Fixed terms have both a defined end and, usually, a fixed rate; some accrue interest to a lump sum at closure, and renewal may trigger a fresh fee. Open-term is more flexible, fixed-term more predictable. Neither is better in the abstract, and the right answer depends on whether you know when you will repay.

What you actually receive, and where you live

Loan proceeds may be dollars or a stablecoin; if you need funds in a bank account, that difference may mean an extra conversion step and cost. And state availability can override every other consideration on this page — several lenders in the comparison do not operate in every US jurisdiction, so check eligibility before doing any arithmetic at all.

Where these numbers come from

Everything in the comparison is drawn from public sources and refreshed on a weekly cadence. We think the sourcing is worth describing plainly, because a comparison is only as good as what sits behind it.

• Each lender’s own published terms are the primary source, and are preferred over third-party reporting wherever both exist. Where a lender’s own pages disagree with one another — which does happen — we use the more specific and more recently updated of the two.
• On-chain and protocol data is used where a lender’s product is funded by a public lending market. That data is often more precise than the lender’s own marketing copy, and it can also supply figures the lender does not publish, such as a liquidation threshold enforced in a contract.
• Variable rates are shown as a short trailing average rather than a spot value. A rate that reprices continuously would otherwise make the comparison an artefact of the moment it was generated.
• Third-party reporting is used sparingly and marked as such. A small number of fields are not published by the lender in any form we can read, because some disclosure pages render only in a browser. Where we rely on trade press for a figure, the table says so.
• Unpublished fields are left empty. A meaningful share of the grid is genuinely not disclosed by anyone. We would rather show a gap than an estimate dressed as a fact.

Lenders are also excluded where no honest comparison is possible. If a lender publishes no rate that a US borrower could actually obtain — for instance where every advertised figure depends on holding its own token — there is no figure to place in the table, and we leave the lender out rather than guess. That does not imply anything about that lender’s competitiveness in either direction.

How to run this comparison yourself

Nothing here requires taking our word for it:

1. Start from your real numbers: the amount you want to borrow, the collateral you hold, the loan-to-value you are comfortable at, and how long you expect to keep the loan.
2. Filter first on the things that disqualify a lender outright — your state, your collateral asset, the loan-to-value you need. Price is only worth comparing among lenders that can actually serve you.
3. For each remaining lender, find the rate for your loan size and loan-to-value, not the advertised floor.
4. List every fee, and establish the basis of each one. “0.20%” of what?
5. Divide collateral-based fees by your loan-to-value, then sum.
6. Compare that total, and weigh it against the structural factors above.

If a lender will not tell you the basis of a fee, that is itself useful information.

Frequently asked questions

Important disclosures

Abra Capital Management, LP (“ACM”) is an SEC-registered investment adviser. Registration does not imply a certain level of skill or training, nor endorsement by the SEC. Abra is one of the lenders described in this article, and readers should assume we have a commercial interest in crypto-backed lending.

Illustrative annualized borrowing cost is calculated as the applicable interest rate plus every fee each lender publicly discloses, expressed as a percentage of the amount borrowed, on the scenario stated with the comparison: a $250,000 loan at 50% loan-to-value held for twelve months. Fees assessed on collateral rather than on the loan are converted at that loan-to-value. Rates are quoted on the basis each lender publishes; some are compounding annual percentage yields and variable, others are nominal rates fixed at origination, and the comparison identifies which is which. Larger or smaller loans fall into different rate tiers at several lenders and produce different figures.

Competitor information is compiled from publicly available disclosures as of the date shown with the comparison and is refreshed weekly. Rates and terms change frequently and without notice. Abra does not warrant the accuracy or completeness of third-party information, and terms actually offered by other providers may differ materially from their published terms. Comparisons are illustrative and are not an offer, solicitation, or recommendation.

Disclaimer. Crypto-backed loans involve substantial risk, including the risk that collateral is liquidated at an unfavourable time. Loans are overcollateralized. Digital-asset prices are volatile. Collateral deployed to DeFi protocols carries additional smart-contract and protocol risk. Abra products and customer accounts are not subject to FDIC or SIPC protection. Nothing in this article is tax, legal, accounting, financial, or investment advice, and nothing here should be relied upon as a recommendation to enter into any borrowing arrangement.