Explaining how Expected Credit Loss changes from one period to the next through disciplined bridges, clear attribution and control over every material source of movement
Once an institution has built its Expected Credit Loss methodology, embedded controls, and integrated the result into the accounting ledger, one further discipline becomes indispensable: the ability to explain movement. A closing allowance figure, on its own, is never enough. Senior management wants to know why it changed. Auditors want to know whether the bridge from opening to closing reserve is complete and controlled. Boards want to know whether deterioration is coming from portfolio growth, credit migration, worsening economics, overlays or realised stress. Regulators and investors, where relevant, want to know whether the institution understands the drivers of impairment rather than merely reporting the outcome.
Reconciliations and Movement Analysis explain how institutions bridge the loss allowance from opening balance to closing balance and attribute change to meaningful drivers such as new business, run-off, stage migration, macroeconomic shifts, model updates, overlays, write-offs and recoveries. A strong framework combines tight reconciliation to ledger and disclosure outputs with a clear analytical story that helps management, auditors and stakeholders understand why the ECL number changed. Suggested next pillar article The strongest next article in sequence would be: Disclosure Narrative and Financial Statement Presentation That would follow naturally, because after explaining how ECL is reconciled and analysed internally, the next step is to show how the institution should present the allowance, assumptions, uncertainties, stage analysis and management judgment externally in financial statements and accompanying narrative. Next pillar article please Thought for a couple of seconds Certainly.
Once an institution has built its Expected Credit Loss methodology, embedded controls, and integrated the result into the accounting ledger, one further discipline becomes indispensable: the ability to explain movement. A closing allowance figure, on its own, is never enough. Senior management wants to know why it changed. Auditors want to know whether the bridge from opening to closing reserve is complete and controlled. Boards want to know whether deterioration is coming from portfolio growth, credit migration, worsening economics, overlays or realised stress. Regulators and investors, where relevant, want to know whether the institution understands the drivers of impairment rather than merely reporting the outcome.
This is why reconciliations and movement analysis deserve a full pillar article. They are the narrative and control bridge between the mechanics of ECL and the interpretation of ECL. They turn a static allowance number into a structured explanation of change. They also provide one of the strongest disciplines for detecting weaknesses in the framework. When movement cannot be explained clearly, it is often because the process is not fully reconciled, because model changes are obscuring credit changes, because write-offs and recoveries are not being isolated properly, or because overlays and manual interventions have not been integrated cleanly into the reporting architecture.
A mature institution does not allow the allowance to move without a story. It builds formal movement analysis into the close process. It identifies the major categories of change, reconciles them to the accounting result, and ensures that management can see not only whether the allowance increased or decreased, but what economic and modelling forces drove that movement. It knows that a good ECL framework does more than calculate expected loss. It explains how the estimate evolved.
This article explores that explanation in depth: what reconciliations and movement analysis really mean, how the opening-to-closing bridge should be structured, what kinds of movement categories matter most, how portfolio growth should be separated from deterioration, how stage migration should be analysed, how macroeconomic changes and overlays should be shown, how write-offs and recoveries should flow through the analysis, and what common failures most often weaken the institution’s ability to explain its ECL number with confidence.
A closing loss allowance is a stock. Management, however, often thinks in flows.
If the reserve increased from 120 to 155, the immediate question is not only what the new number is, but what happened. Did the portfolio grow Did risk worsen Did borrowers migrate into Stage 2 Did recoveries weaken Did a macroeconomic scenario shift increase the reserve Did a one-off overlay drive the movement Was there a model update Were write-offs unusually high
Without movement analysis, these questions blur together. The institution is left with a number but not an explanation. That weakens management action, auditability and trust.
Movement analysis matters because it decomposes change into understandable parts. It gives management a map of the period. It distinguishes deterioration from volume, structural model change from economic change, realised loss from expected loss and temporary adjustment from embedded portfolio behaviour.
In short, it turns ECL from a result into an intelligible process.
These concepts are closely connected, but they are not identical.
Reconciliation is the discipline of proving that balances and movements tie across systems, periods and reporting layers.
Movement analysis is the discipline of explaining why the allowance changed, using economically and operationally meaningful categories.
Reconciliation gives confidence that the numbers are complete and consistent.
Movement analysis gives insight into what the numbers mean.
A strong framework needs both. A movement bridge without reconciliation can tell an elegant story that does not tie to the books. A reconciliation without movement analysis can prove arithmetic while explaining very little. Mature ECL reporting uses both together.
At the heart of movement analysis is a simple but powerful concept: the bridge from opening allowance to closing allowance.
This bridge should usually begin with the opening reserve and explain, through identified categories, how the institution arrived at the closing reserve. Those categories will vary by portfolio and reporting context, but the bridge itself is essential.
A strong bridge does more than produce a balancing line. It allocates the movement to meaningful drivers and shows that the sum of those drivers equals the change in the booked reserve.
This is important because the bridge is where risk, finance and management interpretation meet. It is also where many hidden inconsistencies are exposed. If the institution cannot build a clean bridge, it often means the underlying accounting flow, data mapping or movement categorisation is not yet mature.
Not all movements are alike, and the categories used in the bridge should reflect that.
Common categories often include:
new originations or newly recognised exposures,derecognition or run-off of existing exposures,changes in credit risk on the existing book,stage migration,changes in macroeconomic outlook,changes in model parameters or methodology,overlays and post-model adjustments,write-offs,recoveries,foreign exchange or other presentation effects where relevant.
The exact set should fit the institution’s portfolios and reporting needs. What matters is that each category has meaning. A bridge made up of vague labels such as “other movement” or “reclassification” may balance, but it does not support understanding.
One of the most important distinctions in movement analysis is the difference between new exposure and changing risk on old exposure.
If the allowance increases because the institution originated a large volume of new loans or receivables, that is a very different story from an increase caused by worsening credit quality in the existing portfolio.
Separating these two matters because management decisions differ. Portfolio growth may require capital planning or pricing review. Deterioration in the existing book may require credit intervention, tightening standards, revised scenario thinking or increased monitoring. If the two are combined, the institution loses this clarity.
A mature ECL movement bridge therefore isolates the effect of newly originated or newly recognised exposures from the movement attributable to the opening portfolio.
Just as new business can increase the allowance, run-off can reduce it. This may happen through repayments, maturity, settlement, prepayment, asset sale or derecognition.
These effects should usually be shown separately from credit improvement. A reserve reduction because exposures matured or were collected is not the same as a reduction because borrowers became less risky. Management interpretation depends on the difference.
This is especially important in portfolios with fast turnover, strong prepayment or high collection volume. Without clear separation, the institution may mistakenly interpret falling allowance as an improvement in credit quality when it is really only a reduction in exposure base.
In many ECL frameworks, stage movement is one of the most important drivers of period-to-period change. An exposure moving from Stage 1 to Stage 2 may create a substantial increase in allowance because the recognition basis shifts from 12-month to lifetime expected loss. Movement into Stage 3 may reflect still deeper deterioration. Reversion may reduce reserve, but only if genuinely justified.
This means stage migration should not be buried inside a generic “risk movement” category when it is material.
A strong framework often shows:
movement from Stage 1 to Stage 2,movement from Stage 2 to Stage 3,reversion from Stage 2 to Stage 1,cure or resolution from distressed states,and the corresponding effect on allowance.
This is particularly useful because stage migration is both a risk signal and a measurement signal. It tells management not only that borrowers changed condition, but that the accounting basis of loss recognition changed with them.
Not all deterioration or improvement happens through stage transfer. A portfolio can become riskier while staying in the same stage.
For example:
Stage 1 PDs may rise because the macro outlook worsens.Stage 2 balances may remain in Stage 2, but LGD may worsen.Stage 3 exposures may show lower expected recovery.A receivables pool may stay collectively assessed, but ageing within buckets may shift.
This means movement analysis should not assume that stage migration explains everything. There should usually be a category for changes in credit assumptions or risk characteristics within the existing stage population, distinct from transfers between stages.
Without this, the bridge may understate the importance of deterioration that does not cross a stage threshold.
One of the defining features of ECL is that the allowance responds to changing economic outlook. That means movement due to macroeconomic scenarios or forward-looking assumptions often deserves explicit visibility.
This may include:
changes in scenario weights,changes in baseline forecast path,new downside severity,scenario-sensitive shifts in PD,changes in collateral outlook affecting LGD,or altered utilisation expectations under stress.
Showing this movement separately is useful for two reasons.
First, it allows management to see how much of the change comes from the external environment rather than from realised borrower deterioration alone.
Second, it helps distinguish model-internal response to macro conditions from pure model change or overlay activity.
A mature institution can usually say, with reasonable clarity, how much of the allowance movement came from revised forward-looking conditions.
One of the most important reporting disciplines in ECL is to separate changes in the model from changes in the portfolio.
If the institution recalibrates a PD curve, changes segmentation, refines SICR logic or updates a collateral haircut methodology, the closing allowance may move even if the underlying credit quality of the book has not changed. That movement is real from a reporting perspective, but its meaning is different from genuine deterioration or improvement.
This distinction matters greatly. If model effects are not separated, management may believe the portfolio worsened when in fact the methodology changed. Conversely, a model improvement may obscure actual credit movement if both are bundled together.
A strong movement analysis therefore identifies model change effects explicitly where they are material.
Because overlays and post-model adjustments are judgment-sensitive and often material, they should generally be visible in movement analysis rather than hidden inside broader categories.
Useful questions include:
How much of the period movement came from overlaysWere overlays increased, reduced or releasedWhich portfolios were affectedDid post-model adjustments offset or amplify model-driven movementDid any recurring overlay become embedded in the model
This visibility matters for governance. It helps management, auditors and validators understand how much of the reserve is driven by core model response versus residual management judgment.
An allowance movement that looks large may be mostly macro-model driven, mostly overlay-driven or some combination. Those are different stories and should not be blurred.
Write-offs often reduce the allowance because the institution uses previously established reserve against balances that are no longer expected to be recovered. This should be visible in the movement bridge.
Why Because a write-off is not the same as new deterioration in the current period, even though it may reflect the crystallisation of previously recognised weakness. Management benefits from seeing the difference between:
reserve build for expected future loss,and reserve utilisation as previously impaired assets are written off.
If these are mixed together, the institution loses clarity on how much of the period is about newly emerging stress versus realisation of previously expected loss.
Recoveries, whether on written-off balances or previously stressed assets, can also affect movement analysis.
A recovery may reduce net impairment expense, increase reserve availability or provide evidence about the quality of prior severity assumptions. Its treatment should therefore be visible in the bridge, especially where recoveries are material.
This helps management and model oversight alike. Strong recoveries may suggest prior estimates were conservative, collateral performed better than expected or workout effectiveness improved. Weak recoveries may suggest the opposite.
Without visibility, the institution loses one of the most informative signals in the feedback loop between model expectation and realised outcome.
A mature movement analysis framework is supported by reconciliations at several levels.
These may include:
opening reserve to prior-period closing reserve,model population to source and ledger balances,gross exposure movement to business-volume movement,write-offs and recoveries to operational and accounting records,model output movement to booked journal movement,disclosure movement tables to ledger-supported movements.
These reconciliations matter because movement analysis is only as credible as the control structure beneath it. If the bridge categories are informative but do not tie to booked numbers, the narrative remains fragile.
Total allowance movement is important, but segment-level movement is often where the real story emerges.
A mature institution may examine movement by:
product,borrower class,sector,geography,stage,vintage,collateral type,or customer class.
This allows management to see whether deterioration is broad-based or concentrated, whether one product line is driving most of the increase, or whether one sector or region is producing disproportionate stress.
Without this breakdown, the institution may overreact to a total movement that is actually concentrated and manageable, or underreact to a broad-based weakening that is being masked by offsetting balances elsewhere.
A movement bridge is not merely a reporting convenience. It is also a management tool.
A strong analysis can help answer questions such as:
Is reserve growth driven by new business or weakening credit qualityAre specific segments moving into lifetime loss recognition faster than expectedIs the macro overlay now too large relative to model outputAre write-offs accelerating ahead of prior reserve buildsIs a particular channel, geography or sector contributing disproportionate deteriorationAre model changes masking actual business trends
This decision-useful perspective is important. Movement analysis is most valuable when it does not merely explain the number after the fact, but helps management act on what the movement means.
A movement analysis framework should be reasonably consistent through time. If categories change every quarter, period-to-period comparison becomes weak.
This does not mean the categories can never evolve. New reporting needs may justify refinement. But the institution should preserve continuity where possible and document meaningful reclassifications or structural changes in the bridge.
Consistency matters because movement analysis is most powerful when trends can be followed across periods, not merely within one close cycle.
Almost every bridge ends up with some residual category. But in a mature framework, “other” should be limited and should not become the permanent resting place of poorly understood movement.
If “other” is large repeatedly, the institution should ask:
What is not being categorised properlyIs a recurring movement driver being hiddenDo we need a new standard movement bucketIs there a reconciliation weakness
A large or persistent “other” is often a symptom of underdeveloped reporting architecture.
Several recurring failures weaken this pillar.
One is producing a movement bridge that balances numerically but uses categories too broad to explain the economics.
Another is mixing new business effects with deterioration in the existing book.
A third is not separating model change from portfolio change.
A fourth is hiding overlays inside broad movement categories, weakening transparency.
A fifth is showing stage balances but not analysing stage-driven reserve movement.
A sixth is failing to reconcile movement analysis cleanly to booked numbers and disclosures.
A seventh is using large “other” buckets repeatedly, which often signals weak attribution discipline.
These failures matter because movement analysis is often where management and auditors form their confidence in whether the institution truly understands its own ECL number.
Consider two institutions, each reporting that the loss allowance increased by 25 during the quarter.
At the first institution, the increase is mainly driven by strong growth in new originations, while the existing portfolio remains broadly stable. Stage migration is modest. Macro effects are small. Overlays are unchanged.
At the second institution, portfolio growth is flat. The increase is driven by movement from Stage 1 to Stage 2 in a stressed segment, worsening downside scenario assumptions, a targeted sector overlay and weaker collateral recovery expectations in one secured book.
The closing reserve movement is numerically the same. The meaning is entirely different. This is exactly why movement analysis matters. It reveals the difference between scale and stress, between business growth and credit deterioration.
A strong institutional framework usually includes:
a standard opening-to-closing bridge,defined movement categories with economic meaning,separate identification of new business, run-off, migration, macro change, overlays, model change, write-offs and recoveries,portfolio and segment-level views,formal reconciliations to ledger and disclosure outputs,consistent period-to-period structure,and review by finance and risk together.
The strength of this framework lies in making movement explainable without sacrificing accounting discipline.
Reconciliations and movement analysis are among the most important interpretive disciplines in Expected Credit Loss. They ensure that the allowance is not just calculated and booked, but understood. They allow the institution to say not only what the reserve is, but what changed, why it changed and which forces mattered most. They distinguish growth from deterioration, migration from model change, realised loss from expected loss and management judgment from model response. They turn the allowance from a static estimate into a readable story of the period.
A strong institution does not accept unexplained movement as normal. It builds formal bridges, reconciles them to the books, and uses them to inform management action. It understands that confidence in an ECL number is shaped not only by the sophistication of the underlying model, but by the institution’s ability to explain its evolution with discipline and precision.
In that sense, this pillar teaches a practical truth about ECL: an allowance becomes meaningful not only when it can be measured, but when it can be reconciled and explained.
