By Steve Wofford
For decades, financial institutions have relied on accounting systems as the primary mechanism for evaluating performance. Income statements, balance sheets, and regulatory filings are treated as authoritative representations of profitability because they are standardized, auditable, and governed by well-defined rules.
These characteristics make accounting indispensable for financial reporting, regulatory compliance, and external communication. However, the very features that make accounting reliable also constrain its usefulness for decision-making. Accounting was never intended to measure economic value in a forward-looking sense, and when it is used for that purpose, the results can be materially misleading.
The distinction between accounting and profitability analytics begins with their fundamental objectives. Accounting is designed to record what has already occurred, using a consistent framework that ensures comparability across institutions and time periods.
About Profitability Analytics
Profitability analytics, by contrast, is designed to estimate what is occurring economically, incorporating expectations about future events and behaviors. This includes factors such as expected credit losses, prepayment behavior, funding costs, and capital usage. Because these elements involve uncertainty and require assumptions, profitability analytics is inherently a modeling exercise. The divergence between accounting and economic profitability is therefore not an anomaly but a predictable outcome of their differing purposes.
One of the primary sources of divergence is the treatment of time. Accounting is inherently backward-looking, reflecting realized revenues and expenses based on defined recognition rules. Economic profitability, however, must consider the full lifecycle of an asset or relationship, including expected future cash flows and risks. A loan that appears highly profitable today based on its current spread may be far less attractive when expected losses and prepayments are incorporated.
Limited Usefulness
Similarly, a relationship that appears marginal in current accounting periods may generate significant long-term value through cross-selling or retention. Accounting frameworks are not structured to capture these dynamics, which limits their usefulness for forward-looking decision-making.
Funding represents another critical area where accounting and economic perspectives diverge. Accounting measures interest expense based on the institution’s booked rates, which reflect historical funding decisions rather than current market conditions. Profitability analytics introduces the concept of Funds Transfer Pricing (FTP), which assigns a cost of funds based on market rates and the duration or behavior of the asset or liability.
This adjustment reflects the true economic cost of deploying capital and managing interest rate risk. Without FTP, profitability measures can be significantly distorted, as they fail to account for the opportunity cost of funds. In financial institutions, this omission is not a minor limitation but a structural deficiency that undermines the validity of profitability analysis.
Further Amplification
Risk further amplifies the divergence between accounting and economic profitability. Accounting recognizes losses based on defined triggers or expected loss frameworks such as CECL, but these are still governed by prescribed rules and timing conventions. Economic profitability, on the other hand, incorporates risk continuously through measures such as probability of default, loss given default, and exposure at default. This allows for a more accurate representation of the economic cost of risk over time.
As a result, accounting profitability often appears stable until losses are realized, while economic profitability may indicate deterioration well in advance. This difference is critical for institutions seeking to proactively manage risk rather than react to it after the fact.
Additional Limitations
Cost attribution also illustrates the limitations of accounting-based approaches. In accounting systems, expenses are recorded where they occur organizationally, often at a departmental or functional level. This structure does not necessarily reflect the underlying drivers of those costs or how they relate to specific products, customers, or activities. Profitability analytics seeks to address this through methodologies such as Time-Driven Activity-Based Costing, which assigns costs based on the resources consumed to deliver a product or service.
By focusing on causality rather than organizational structure, this approach provides a more accurate view of economic cost. While it introduces additional complexity, it also enables more informed decisions about pricing, product design, and resource allocation.
Additional Shortfall
Capital allocation represents another area where accounting falls short as a measure of economic performance. Accounting systems do not assign capital to individual instruments or relationships in a way that reflects their risk profile. Economic frameworks such as Risk-Adjusted Return on Capital address this by recognizing that capital has a cost and must be allocated based on the level of risk being assumed.
This allows institutions to evaluate whether a given activity is generating sufficient return relative to the capital it consumes. Without this adjustment, institutions may inadvertently prioritize growth in low-return or high-risk areas, diluting overall performance. The absence of capital allocation in accounting-based profitability measures is therefore a significant limitation.
A Common Objection
A common objection to advanced profitability analytics is that it introduces model risk due to its reliance on assumptions and estimates. While this concern is valid, it is often misunderstood. All profitability systems involve some degree of modeling, including those that rely primarily on accounting data. The difference is that accounting-based approaches embed assumptions implicitly, while economic models make them explicit.
Choosing not to model key factors such as funding costs, risk, or customer behavior does not eliminate assumptions; it simply obscures them. In many cases, the risk associated with omitting these factors is greater than the risk of modeling them imperfectly.
A Frequent Source of Tension
The challenge of reconciling profitability analytics with accounting results is a frequent source of tension within organizations. Users often expect profitability measures to tie directly to the general ledger, viewing discrepancies as errors or inconsistencies. In reality, these differences reflect the distinct objectives of the two systems.
Accounting provides a record of realized financial performance, while profitability analytics provides an estimate of economic value. Attempting to force alignment between the two can compromise the integrity of the economic model, reducing it to a restatement of accounting results. A more effective approach is to recognize the complementary roles of each system and use them accordingly.
The implications of this distinction are significant for decision-making. Pricing decisions based solely on accounting measures may fail to account for funding costs, risk, and capital requirements, leading to suboptimal outcomes. Similarly, strategic decisions based on historical profitability may overlook emerging trends or future risks.
The Framework
Profitability analytics provides a framework for incorporating these considerations into decision-making, enabling institutions to align their actions with economic reality. While this approach requires more sophisticated data and modeling, it also offers a more accurate basis for evaluating performance and guiding strategy.
Accounting will always play a critical role in financial institutions, providing the foundation for reporting, compliance, and external communication. However, it is not sufficient for managing economic performance in a complex and dynamic environment. Profitability analytics fills this gap by extending beyond the constraints of accounting to incorporate forward-looking, risk-adjusted, and behaviorally informed perspectives.
The choice is not between accounting and profitability analytics, but between relying solely on historical records or augmenting them with economic insight.
The Ultimate Question
Ultimately, the question is not whether profitability analytics introduces model risk, but whether institutions are willing to make decisions based on explicit, economically grounded assumptions rather than implicit and incomplete ones. Accounting will continue to provide an essential record of what has occurred, but it cannot, on its own, explain why it occurred or what is likely to happen next.
For institutions seeking to understand and improve their true economic performance, moving beyond the general ledger is not optional. It is a necessary step toward more informed and effective decision-making.
Steve Wofford is CEO of Kohl Analytics Group.
