ERC-20 allowance is the amount of tokens an owner authorises a spender to move from their account on an Ethereum token contract. The contract tracks the approved limit separately from the balance, and the spender can use that allowance through approved calls until it is reduced or exhausted.
What ERC-20 Allowance Means in Practice
ERC-20 allowance is a delegated spending limit, not a transfer of ownership. It lets one address act on another address’s behalf within a bounded token amount, which is why allowance is central to token approvals and contract-based spending.
The key security property is that the allowance exists separately from the wallet balance. A spender can only move tokens that the owner has explicitly approved, but that approval can still create meaningful exposure if the limit is too large or left in place longer than intended.
How Allowances Work in Token Contracts
Allowance is usually established through an approval call that records how much a spender may use. The spender then consumes that approved amount through later contract calls, and the remaining allowance is reduced as those calls succeed.
This model supports decentralised applications, exchanges, and automated contract workflows, but it also introduces a trust dependency on the spender contract. The owner is not simply sharing a balance, they are authorising a specific contract address to act within defined bounds.
Because the contract tracks approval state independently, allowance can persist after the original interaction is over. That persistence is useful for convenience, but it also means old approvals may remain active if they are not reviewed or cleared.
Security Implications of Token Approval
Allowance creates a controlled delegation path, which is powerful but sensitive. If a spender contract is compromised, malicious, or later upgraded in an unsafe way, the approved limit can become a ready-made path for token extraction within the approved ceiling.
Allowance also matters because users often approve more than is strictly necessary to avoid repeated confirmations. That convenience trade-off can turn a single approval into a larger exposure window than the user intended.
In practice, the relevant security question is not whether an approval exists, but whether the approved spender is still trusted, still needed, and limited to the smallest workable amount.
Common Failure Modes and Safer Approval Patterns
Allowance failures usually come from excess privilege, stale approvals, or poor spender selection. A high or unlimited approval can make a later compromise far more damaging than a narrow, time-bounded approval would be.
Safer patterns include granting only the amount needed for the immediate action, reviewing old approvals periodically, and revoking allowances that are no longer required. Where a workflow can function with repeated small approvals, that is often safer than leaving a long-lived broad approval in place.
Users should also distinguish between the token contract they intend to trust and any frontend, router, or aggregator that asks for approval. The approval should match the actual contract that will spend the tokens, not just the site requesting consent.
Risk and Threat Considerations
ERC-20 allowances can be abused when a spender contract, approval workflow, or user interface is deceptive or compromised. The main risk is that a previously trusted approval becomes an active token-drain path without requiring the victim to sign a fresh transfer for every movement.
Failure mechanism: An attacker exploits an overbroad, stale, or misleading approval and uses the authorised spender path to move tokens up to the remaining allowance.
Impact: Token loss can occur quickly, and the damage is often limited only by the size of the approved allowance rather than by the current wallet balance or by any later user action.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
OWASP API Security Top 10 and OWASP Non-Human Identity Top 10 address the attack and risk surface, while NIST SP 800-53 Rev 5 sets the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| OWASP API Security Top 10 | API5 — Broken Function Level Authorization | Allowance is an authorization boundary on contract spending |
| Recommendation — Constrain token-spending functions to the minimum approved contract path. | ||
| NIST SP 800-53 Rev 5 | AC-6 — Least Privilege | Approvals grant bounded delegated access and should be minimized |
| IA-5 — Authenticator Management | Allowance state behaves like managed access material requiring lifecycle control | |
| Recommendation — Grant the smallest viable spending authority and revoke obsolete approvals. Review and rotate approval pathways when delegated access is no longer needed. | ||
| OWASP Non-Human Identity Top 10 | NHI-05 — Overprivileged NHI | Contract approvals can grant excessive token-moving authority |
| NHI-07 — Long-Lived Secrets | Persistent approvals create enduring delegated access exposure | |
| Recommendation — Limit contract allowances so a compromised spender cannot drain excess tokens. Expire or revoke stale allowances instead of leaving them in place indefinitely. | ||
Practitioner Guidance
Why practitioners should care: Allowance design is a practical authorisation decision, not just a wallet feature. Teams building token-enabled products should treat approvals as delegated privilege and minimise how much authority each approval actually grants.
What to watch for: Large, unlimited, or long-lived approvals deserve review because they create avoidable exposure. Users and product teams should pay particular attention when a contract requests more authority than the immediate transaction appears to require.
Practitioner takeaway: The safest allowance is the smallest one that still lets the intended contract complete the intended action.
Related resources from NHI Mgmt Group
Deepen Your Knowledge
Free weekly newsletter
Subscribe to the NHI & AI Identity Journal
The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.
Bonus 33% off our NHI Course when you subscribe.
Reviewed and updated by the NHIMG editorial team on October 29, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org