A contract for difference is exactly what the name says. Two parties agree to settle the change in an instrument's price between the moment the contract opens and the moment it closes. If the price rises, the seller pays the buyer the difference. If it falls, the buyer pays. Nothing is delivered, nothing is settled at a custodian, and no share certificate exists.
People conflate CFDs with investing because the screen looks the same. The ticker is the same, the chart is the same, the buy button is in the same place. What differs is the legal object you end up holding and, more importantly for the account, the cost of holding it.
Ownership, and what disappears without it
Owning shares gives you the residual rights that come with the asset: voting at the annual meeting, dividends paid in cash, participation in corporate actions, and a holding that persists if your broker fails, because it is held for you rather than by you.
A CFD gives none of that. Brokers replicate the economic effect of dividends by crediting long positions and debiting short positions with an adjustment on the ex-dividend date, but that is a contractual payment from the broker, not a distribution from the company. Voting rights do not exist. Corporate actions are handled by adjusting the contract rather than by giving you a choice.
The counterparty question is the one that gets underweighted. Your CFD position is an obligation of the broker, so the broker's solvency and regulatory regime is part of the trade. Rules on client fund segregation and, in some jurisdictions, investor compensation schemes determine what happens to your money if the firm fails. Those protections vary enormously between a firm authorised in a major jurisdiction and one licensed offshore.
Costs that diverge as the calendar runs
This is the part that decides which product suits a given trade, and it is straightforward arithmetic.
| Cost | CFD position | Owned shares |
|---|---|---|
| Entry and exit | Spread, sometimes commission | Commission, sometimes stamp duty or transaction tax |
| Holding overnight | Financing charged nightly on the full notional | Nothing, beyond any custody fee |
| Income | Dividend adjustment, credited or debited | Dividend paid, subject to withholding |
| Currency | Conversion applied to profit and loss | Conversion applied at purchase and sale |
| Capital required | Margin, a fraction of the exposure | Full purchase price |
The line that matters is the second one. Financing accrues on the whole notional value of the position, not on the margin you posted. Hold a 50,000 unit exposure for eight months and the accumulated charge becomes a meaningful percentage of that notional, deducted quietly from equity every night. For an intraday or multi-day position it is close to irrelevant; for a long-term holding it works directly against you. The mechanics are identical to the swap charged on currency positions.
If your intended holding period is measured in years, a leveraged derivative with a nightly financing charge is the wrong container for the idea. That is not a criticism of CFDs. It is a statement about what they were built for.
Leverage is the whole point, and the whole risk
The reason CFDs exist for retail traders is margin. A few percent of the exposure is enough to open the position, which frees capital and allows small accounts to take positions they could not otherwise fund. It also means a modest adverse move can consume the entire deposit, and unlike owned shares there is a stop out level at which the broker closes you.
Regulators have taken a view on this. Retail leverage caps in the European Union and the United Kingdom limit how much exposure a retail client can take per asset class, with tighter limits on more volatile instruments, and negative balance protection stops a retail account from going below zero in those jurisdictions. The background to those measures is set out in the piece on leverage caps. Offshore firms often offer far higher ratios, and the trader carries the difference in risk.
An owned share portfolio can fall a long way and still be held. A leveraged CFD on the same fall gets closed out before any recovery arrives. The instrument changes the outcome of an identical market view, which is why understanding leverage comes before choosing between the two.
What CFDs do that ownership cannot
Shorting is the honest advantage. Selling short in a custody account requires borrowing the asset, which is fiddly, sometimes unavailable and priced by scarcity. Selling a CFD is one click in either direction with identical mechanics. For anyone whose method takes both sides of a market, that symmetry is worth a great deal.
Access is the second. One CFD account reaches currencies, indices, commodities and crypto exposure in the same platform with the same margin pool, without opening accounts at four different venues. Position sizing is finer too: fractional lots let a small account take a properly sized position in an instrument whose underlying unit would be unaffordable.
Hedging is the third, and it is the reason CFDs originally existed. A holder of a real portfolio can offset exposure temporarily without selling the underlying holdings and realising a tax event or losing a position they intend to keep.
Choosing the container for the idea
My rule of thumb is horizon first, instrument second. If the idea plays out inside a few weeks and needs leverage or the short side, a CFD fits. If the idea is measured in years and rests on the asset compounding, ownership fits, and the financing charge alone should settle the argument.
Availability sometimes decides for you. CFDs are not offered to retail clients in the United States, and several other jurisdictions restrict or prohibit them, so the choice may not exist depending on where you live. Where both exist, treat them as separate accounts with separate purposes rather than as two doors into the same activity. Mixing them, and especially funding a long-term view through a leveraged contract, is how people end up stopped out of a position they were right about. Trading CFDs carries a high risk of loss, and a large share of retail accounts lose money on them.
"A CFD is a rented view on a price. Renting is fine. Just do not confuse it with owning, and do not rent for ten years."
— Alex Onta, Executive Director, SINGUARD
Key Takeaways
- A CFD is a contract with the broker, so broker solvency and regulation are part of the position.
- Overnight financing is charged on the full notional value, which penalises long holding periods.
- Margin can close a leveraged position that an owned holding would have survived.
- Easy shorting, one margin pool across asset classes and fine position sizing are the real advantages.
Frequently Asked Questions
Do I own anything when I buy a CFD?
No. A contract for difference is an agreement with the broker to exchange the change in an instrument's price between opening and closing. No shares are bought, nothing is registered in your name, and there is no custodian holding an asset for you. What you hold is a claim against the broker, which is why the financial strength and regulatory status of that broker matters more than it would for a custody account.
Are CFDs cheaper than buying shares?
For a position held hours or days, often yes, because entry costs are low and no settlement or custody fees apply. For a position held months, usually no, because an overnight financing charge accrues every night on the full notional value rather than on the money you deposited. The break-even point depends on the financing rate and the entry costs, and it is worth calculating before committing to a long hold.
Can I hold a CFD forever?
Cash CFDs have no expiry, so technically yes, but the financing charge accumulates for as long as the position is open and it is deducted from account equity. On a multi-year horizon that cost can become a large fraction of the notional value. CFDs are built for shorter holding periods, and using one as a substitute for a long-term holding works against the design of the product.