Intercompany reconciliation is the process of agreeing balances and transactions between entities inside the same group, so that they cancel exactly when the consolidated statements are prepared. If entity A says it is owed 100 and entity B says it owes 95, the group cannot report either figure honestly until the 5 is explained.
​

The work exists because elimination is not optional. Group accounts must present the organization as one economic entity, which means internal dealings have to disappear entirely. They only disappear cleanly if both sides recorded them identically, and the reconciliation is what establishes that they did.
​

It is the matching of each intercompany balance and transaction against its counterpart in the other entity's books, followed by resolution of any difference. The output is a set of agreed positions that eliminate to zero in consolidation, plus an explained list of anything that does not yet agree.
​

The requirement comes from the accounting standards rather than from internal preference. In preparing consolidated financial statements, intra-entity balances and transactions shall be eliminated under ASC 810, and any intra-entity profit or loss on assets remaining within the group is eliminated with it.
​

The international requirement is the same in substance. IFRS 10 requires elimination of intra-group balances, transactions, income and expenses, which is why the discipline applies regardless of which framework a group reports under.
​

Five causes account for most differences: timing, where one side has recorded and the other has not; currency, where each side translated at a different rate; disputes over amount or existence; transactions posted to the wrong counterparty; and unilateral entries such as a management charge one side never accepted.
​

Timing differences dominate by volume and are the least interesting. A charge posted on the last day of a period and received the following day appears as a difference on both sides while being entirely correct, and it should be identified as in-transit rather than investigated.
​

Currency differences are structural rather than errors, arising because two entities with different functional currencies translate the same transaction independently. They need an agreed convention about which rate governs, not a search for a mistake.
​

Disputes and unilateral entries are the substantive categories. These are genuine disagreements between two parts of the same organization, and they resolve through a decision rather than a reconciliation rule. They matter disproportionately because ASC 810 requires elimination in full, so an unagreed balance cannot simply be carried forward into the group accounts.
​

The table separates what actually appears in an intercompany queue by cause, since the resolution path differs sharply. Treating a timing difference and a disputed charge as the same kind of item is the most common reason intercompany close work expands to fill the available days.
​

Only three of the six require a human decision. The rest are rule-resolvable, and groups that close quickly are usually the ones that resolved the rule-resolvable categories by rule rather than by correspondence between finance teams.
​

When one entity sells to another at a margin and the buying entity still holds the asset at period end, that margin is profit the group has not actually earned from anyone outside it. It must be removed from the consolidated position, because no external transaction has occurred.
​

The standard is specific about the mechanics. Intercompany income should be eliminated from the applicable asset in the consolidated balance sheet on a before-tax basis, and the concept usually applied is gross profit or loss.
​

A detail that catches groups out: the amount eliminated is not reduced by the presence of outside shareholders in a subsidiary. Under ASC 810 the amount of intra-entity income to be eliminated is not affected by the existence of a noncontrolling interest.
​

Agree balances before close rather than during it. A group that starts reconciling once the period has ended has placed a negotiation between two finance teams on the critical path of its reporting timetable, which is why intercompany work so often becomes the binding constraint on close speed.
​

The practical pattern is a cut-off several days before period end, with balances exchanged, matched, and disputed items escalated while there is still time to post corrections. Anything unresolved at the deadline goes to a documented policy rather than to a discussion. The reporting logic is the same one a 2006 Journal of Accountancy analysis applied to reconciliation generally: finish early enough that adjustments feed the reported numbers instead of trailing them.
​

That policy is what prevents deadlock. A rule stating that unresolved differences below a threshold are absorbed by a designated entity, and above it are escalated to a named decision maker, converts an open-ended argument into a bounded one.
​

The originating entity should own it by default, because it holds the underlying documentation. Where a difference is genuinely disputed, ownership passes to a group-level owner with authority to decide, since two entities of equal standing have no mechanism to resolve a disagreement between themselves.
​

Ownership rules matter more here than in most reconciliations. External reconciliation has a counterparty with an incentive to engage; intercompany reconciliation has two internal teams with competing period results, and without an escalation path the difference simply persists.
​

Segregation still applies within the process. GAO's internal control standards hold that key duties be divided among different people, which in a group context means the entity raising a charge should not also be the entity that unilaterally decides the dispute in its own favour.
​

Prevent at origination rather than detect at close. A charge posted through a process that notifies the receiving entity, uses an agreed counterparty code, applies a group rate convention, and requires acceptance above a threshold will rarely become a reconciliation item at all.
​

Counterparty coding is the highest-yield control. A transaction posted against the wrong entity breaks two balances simultaneously and is invisible on the originating side, so validating the counterparty at entry prevents a class of difference that is expensive to trace afterward.
​

Acceptance workflow handles the unilateral charge problem directly. If a management fee or cost allocation requires the receiving entity to accept it before it posts, the disagreement surfaces when it arises rather than at close.
​

Engines built for this reconcile AR, AP, bank, PSP, and intercompany data through one rule-based engine, which matters because intercompany items frequently originate in the same systems as external ones.
​

Bank reconciliation compares your records against an external party who has no stake in your result and whose statement is authoritative. Intercompany reconciliation compares two internal records of equal standing, neither authoritative, both belonging to teams whose reported results move in opposite directions when a difference is resolved.
​

That symmetry is what makes it harder. With a bank there is a right answer to discover; with an intercompany difference there is often a decision to make, and the two entities may reasonably disagree about which of them should absorb it.
​

The consequence of leaving it unresolved also differs. An unreconciled bank item affects one entity's cash position. An unreconciled intercompany item fails to eliminate and therefore misstates the consolidated group, which is a reporting error rather than a local one.
​

The governance answer is an escalation path with a named decision maker, reinforced by the general control principle that key duties be divided among different people so no entity adjudicates its own dispute.
​

Internal loans produce two balances that must agree and interest that must agree on both sides, then eliminate entirely. A parent lending to a subsidiary creates a receivable and a payable of equal size that cancel in consolidation, leaving nothing in the group statements.
​

Interest is where these commonly drift. The two entities may accrue on different day-count conventions or at different points in the period, producing a difference that recurs every month and is genuinely small each time while never resolving.
​

The prevention is a single amortization schedule maintained centrally and used by both sides, rather than each entity computing its own from the loan terms. One schedule cannot disagree with itself, and it also gives the auditor a single artifact to test rather than two computations to compare.
​

The elimination requirement is unambiguous regardless of structure. IFRS 10 requires intra-group balances, transactions, income and expenses to be eliminated in full, and ASC 810 requires the same for intra-entity balances and transactions.
​

Each side needs its own ledger detail, a shared counterparty code, the supporting document for every charge, and an agreed rate table where currencies differ. Without a shared counterparty code the two ledgers cannot be joined at all, which is why coding discipline precedes any tooling decision.
​

The supporting document is what converts a dispute into a decision. Two entities disagreeing about a management fee with no underlying agreement to point at will keep disagreeing, because there is nothing to test the claim against.
​

Rate tables need to be published centrally rather than chosen locally. When each entity selects its own rate source for the same internal transaction, the resulting difference is guaranteed and recurs every period.
​

Systems that reconcile AR, AP, bank, PSP, and intercompany data through one rule-based engine hold these records together, which matters because intercompany items usually originate in the same subledgers as external ones and are only distinguishable by counterparty.
​

Track the number of open differences, their aggregate absolute value, their age profile, and how many required a decision rather than clearing by rule. Absolute value matters more than net, because two offsetting differences net to nothing while representing two unresolved errors.
​

The age profile is the leading indicator. A stable count with a rising oldest bucket means easy items are clearing while hard ones accumulate, and the hard ones are the disputes that will eventually reach a threshold where they must be decided under pressure.
​

Decision count is the metric that tells you whether prevention is working. A queue that clears mostly by rule is functioning; one that clears mostly by correspondence between finance teams will not survive growth in entity count, because the number of entity pairs grows faster than the number of entities.
​

These measures are also what an auditor will ask to see, since under AS 2201 the question is whether the control operates, not whether a spreadsheet exists.
​

Report by entity pair rather than in aggregate. Intercompany problems concentrate in specific relationships, usually where one entity charges another for shared services, and a group-level figure conceals which pair needs attention.
​

Different ledgers multiply the reconciliation work, because balances must be extracted, normalized, and mapped before they can be compared. Chart of accounts differences, entity naming differences, and posting convention differences all have to be resolved in a translation layer that then becomes something to maintain.
​

The pragmatic answer for most groups is not a single ledger but a single intercompany protocol: one counterparty coding scheme, one rate convention, one cut-off calendar, one acceptance workflow. Those four can be imposed across heterogeneous systems without replacing any of them.
​

Acquisitions are where this discipline is usually tested. A newly purchased subsidiary arrives with its own conventions, and the period before it adopts the group protocol reliably produces the largest intercompany differences the group will see that year.
​

Budgeting for that period explicitly is more effective than treating it as a surprise. The integration work that reduces intercompany differences is counterparty mapping, rate alignment, and cut-off alignment, and none of it requires migrating that subsidiary onto the group ledger first.
​

Whatever the system landscape, the reporting obligation does not flex. Intra-entity balances and transactions shall be eliminated in consolidation, and system heterogeneity is an implementation problem rather than an exemption.
​

Intercompany balances are related-party by definition and eliminate against each other, which makes them a place where errors can offset and hide. An unreconciled intercompany position is also a direct threat to the consolidated numbers, since anything that fails to eliminate lands in the group result.
​

The consequence is defined by standard. Under PCAOB AS 2201, a material misstatement that the company's own internal control did not first detect is a strong indicator of a material weakness, and intercompany reconciliation is the control expected to detect misstatements in these balances.
​

A 2006 Journal of Accountancy analysis made the general case, urging that accounts be risk-rated and higher-risk ones reconciled early enough for adjustments to reach the ledger. Intercompany accounts rank high on that scale in any multi-entity group.
​

Balances agree before the period closes, not during it. Differences are classified by cause with an owner and a deadline, in-transit items age out by rule, disputes escalate automatically at a threshold, and the group can state its unreconciled intercompany position at any point rather than only at close.
​

The strongest signal of a working process is that intercompany is not on the critical path. When agreement happens ahead of close, the consolidation step becomes mechanical, and the group stops discovering disagreements at the moment it has least time to resolve them.
​

Prevention should be visibly reducing the queue over time. If the same entity pair generates the same category of difference every period, the process is detecting a defect it is not fixing, and the fix belongs at origination rather than in the reconciliation.
​

Finally, the elimination itself should be reproducible. Someone should be able to show which balances eliminated, at what amount, and on what basis, including that intercompany income was removed on a before-tax basis using gross profit or loss as the concept applied.
​

Faster settlement removes in-transit differences, which are the largest category by count in most intercompany queues. When an internal transfer settles and confirms immediately, both entities can record it in the same period rather than straddling a cut-off, and the timing category largely disappears from the queue.
​

It does not touch the substantive categories. A disputed management charge, a wrong counterparty code, or unrealized profit on inventory held inside the group are unaffected by how fast money moves, because none of them are caused by settlement timing. The elimination requirement under IFRS 10 is indifferent to how quickly cash moves; it concerns whether both sides recorded the same transaction.
​

Cross-border internal transfers are where the effect is largest, since these carry the longest settlement windows. Domestic rails have compressed this substantially already, with Nacha reporting Same Day ACH volume growing 16.7 percent in 2025 to 1.45 billion payments worth $3.92 trillion. The stablecoin market held $311.0 billion in total supply as of September 9, 2026, led by USDT at $183.4 billion and USDC at $74.4 billion, according to DeFiLlama.
​

Eco operates the routing and execution layer that stablecoin value moves through, so an internal transfer between group entities produces one settlement record both sides can reference, rather than two systems recording the same movement independently and later disagreeing about its timing.
​

That addresses the in-transit category and nothing else. Elimination entries, unrealized profit calculations, counterparty coding, and dispute resolution remain accounting work that no settlement layer performs. Eco is not a consolidation tool and does not produce group accounts.
​

For the record layer, Eco's comparison of stablecoin settlement APIs with audit trails covers what a settlement record should expose.
​

Methodology. Elimination requirements, before-tax treatment of intercompany income, and the noncontrolling interest point are from PwC Viewpoint's summary of ASC 810 intercompany transactions, retrieved September 9, 2026, which cites ASC 810-10-45-1 and related paragraphs. The international requirement is per IFRS 10. Audit consequences are per PCAOB AS 2201. This article describes general requirements and is not accounting advice for a specific group structure. Stablecoin supply is from DeFiLlama as of September 9, 2026 and moves intraday.
​