The tyranny of metrics

We live in an age of big data, where credit analysis has largely become an exercise in evaluating vast digital mosaics: loan to value and debt-to-EBITDA ratios, probabilities of default, expected losses, FICO scores, cash flow waterfalls, and downside scenarios. As technology has facilitated both more timely and plentiful information, lenders have never had such granular visibility into the financial condition of borrowers. With seemingly infinite data at our fingertips – and artificial intelligence (AI) tools at our disposal to accelerate the underwriting process – credit risk should be lower than ever; yet, amid this abundance of information, an essential aspect of credit underwriting may have quietly faded into the annals of history. Could it make a comeback?

The extension of credit, at its core, is not purely a spreadsheet exercise. It is an act of belief.

The Latin root of the word – credo, “I believe” – is a helpful starting point. Credit is a human relationship, one grounded in trust. Rigorous financial analysis must inform and underpin that relationship, sharpen it, and occasionally signal to the credit provider to run the other way. But data cannot fully replace human instinct. As several recent examples show, the danger of modern underwriting is not that we use too much information, but that the comfort of information tempts us to abandon the harder work of judgment – the ability to assess the management integrity, governance practices, and economic incentives of imperfect human beings.

Thoughtful credit underwriting requires lenders to construct downside financial models with thousands of Excel cells and dozens of potential future scenarios. Those who fail to do comprehensive, fundamental credit research will underperform. But the output of these models can at times offer the illusion of safety. Occasionally, the market reminds us that deep personal relationships, firsthand knowledge, and qualitative assessment are not just relics of a less sophisticated era. As history suggests, credit cycles have occurred consistently across centuries, and they tend to follow a similar arc:

    1. Capital becomes plentiful

    2. Competition intensifies

    3. Pressure to scale accelerates

    4. Risk premiums compress as more capital chases fewer attractive opportunities

    5. Caution fades

    6. Standards for select market participants quietly erode

Case in point: Since 2023, measured investor demand has remained strong even as measurable supply has lagged. This creates a competitive deployment environment in which discipline carries the greatest opportunity cost – and therefore, is hard to maintain. Thus, the risk is not always imprudence; more often, it arrives through gradual concessions made to preserve competitiveness.

Whether it’s the proliferation of nonbank lending or the historically tight spreads in corporate credit, today’s market’s confidence is visible in both price and structure. U.S. corporate bond spreads reflect a market that is priced for perfection.

Structurally, the gradual erosion of lender protections reveals a borrower-friendly marketplace. In 2006, covenant-lite loans – structures that remove the financial maintenance tests that historically served as lenders’ early warning mechanisms – accounted for less than 1% of the total U.S. leveraged loan market; by 2026, this number rose to 91.53%. Alongside this shift, market norms have evolved in ways that further favor borrowers, including the routine use of liability management exercises¹ that reflect the increased flexibility in modern capital structures. Such conditions do not imply imminent disruption. However, if history is any guide, the tide always turns as lenders stretch too far – spreads widen, markdowns occur, and capital is permanently lost. Then, the cycle begins anew as bargain hunters swoop in to purchase discounted assets, providing a floor to values and ushering in a new cycle. These cycles are not over. In an environment where risk is priced near perfection and cracks are beginning to emerge, it is worth revisiting some of the first principles of credit underwriting.

Remembering the five Cs of credit

Recent events should feel uncomfortably familiar. The well publicized fraud events of the past year and signs of rising volatility serve as yet another reminder of an old French adage: “Plus ça change, plus c’est la même chose.” The more things change, the more they remain the same. We seem to be entering another period where history rhymes.

In this period, it may be helpful to revisit the old-fashioned but tried-and-true mnemonic device: the five Cs.

Whether financing a manufacturing company’s strategic acquisition, backing a shipment of goods across the ocean, or purchasing a complex structured note, the foundational principles of underwriting endure. The five Cs of credit remain as relevant today as they were generations ago:

• Character: The will to pay
• Capacity: The ability to pay
• Capital: The borrower’s own “skin in the game”
• Collateral: The secondary source of repayment
• Conditions: The broader economic and industry environment

While modern markets devote extraordinary attention to four of these five Cs, character – the first C – has become increasingly marginalized in a world of financial innovation, AI, increased intermediation, and big data. "Character” refers to a borrower’s track record of business transparency, integrity in financial dealings, and willingness to repay creditors, rather than any assessment of political views, personal views, or lawful industry participation. Given the time required to evaluate it and the challenge of measuring it precisely, character has been treated as subjective, unquantifiable, or even quaint. This is not entirely surprising. The other four Cs are fundamentally problems of financial risk, yielding to data, stress scenarios, and models. The measurement of character, by contrast, is inherently imprecise.

Risk factors that cannot be measured are far harder to build into a scalable, repeatable process. The assessment of character resists the spreadsheet. Over time, especially in periods of sporty growth, markets tend to de-emphasize it, not because participants are blind to its importance, but because it is harder to embed in a process. Judgment about character, even when correct, is more difficult to leverage.

The result, perhaps, in our current environment has not been so much the disappearance of qualitative judgment altogether, but rather a gradual repricing of its importance. Standardized diligence and background checks are indispensable, and in many cases, they improve underwriting discipline. However, they can also create the impression that what is hardest to quantify – the complex realities of human behavior – can be reduced to a repetitive task. The COVID-19 period of less travel and Zoom-only meetings exacerbated this trend, as fewer investment professionals felt it necessary to walk the factory floor and look into a borrower’s eyes.

A CEO with strong character cannot protect a creditor from a significant industry disruption at a highly levered enterprise – but at the margin, no assessment of a management team’s talent is complete without it. And in periods when leverage levels are elevated and losses begin to rise, those creditors who endure the hard work of evaluating character are more likely to succeed.

Financial history offers some useful reminders. Consider the Brown Brothers’ “Circular Letter of Credit,” the 19th century precursor to the modern credit card. As The New York Times once observed, its effectiveness rested on its “confidential character,” requiring the firm to be “thoroughly acquainted with the standing of the persons” involved – knowledge that extended well beyond cash balances or holdings of “sound stocks.”

A century later, J.P. Morgan made the same point. Testifying before Congress in 1912, he argued: “The first thing [credit is based upon] is character. Before money or anything else … because a man I do not trust could not get money from me on all the bonds in Christendom.” Capacity might tell you whether a borrower can pay. Only character tells you whether they will.

That insight has gone in and out of fashion, and the number of devotees to the five Cs has gradually declined, nearly to the point of extinction during the pandemic.

What does a healthy credit process look like?

Today’s credit processes are increasingly focused on scale and efficiency. They understandably gravitate toward key metrics and performance indicators: debt-to-EBITDA, debt to enterprise value (the backbone of the loan-to-value ratio in corporate credit), and cash flow available for debt service. These metrics are fundamental aspects of a borrower’s capacity to repay, and any credit analysis without them is insufficient. But they can also become a trap, offering a false sense of precision and certainty that history repeatedly exposes itself as fragile. Models assume continuity. Markets specialize in rupture. Character often only comes to light in the darkest of markets. Are we at a similar inflection point?

Another example of the market’s gradual loss of discipline is the widespread growth of EBITDA addbacks in leveraged finance and direct lending. Despite tightening spreads, these addbacks have increasingly inflated covenant-defined earnings, obscured true leverage, and reduced equity cushions for lenders. While certain legitimate one-time addbacks should not cause concern, the risk of nonrecurring addbacks becoming permanent has increased, and the additions of addbacks stemming from mergers and acquisitions (M&A) synergies and other future events should be scrutinized. In the future, lenders may consider capping addbacks as a percent of EBITDA.

That is what makes the present moment so interesting. By many macroeconomic measures, the backdrop appears broadly supportive: Unemployment was 4.3% in January, real gross domestic product (GDP) grew 2.2% in 2025, and corporate earnings have generally beat expectations. Yet, average leverage in the leveraged loan market sits around 5x debt-to-EBITDA and has not come down with earnings growth. This is elevated enough that repayment capacity remains heavily dependent on earnings durability and refinancing access. In other words, the burden has not reset – conditions have merely remained favorable enough to carry it.

Analysis of collateral remains equally important. In the event of a default, how do I access any collateral that may support my exposure? Or mitigate the risk that it could be pledged to multiple lenders? Several recent examples testify to the importance of the “trust but verify” approach to collateral management – and the importance of knowing who’s on the other side of the transaction.

Focusing on the borrower’s capital – the “skin in the game” of key executives and owners – is equally critical. The lender’s alignment with management and ownership is especially important. For example, what is the incentive for an equity sponsor to support a company whose earnings are declining if it has already generated a 2x multiple on invested cash through three dividend recaps over eight years?

And, as the recent events in technology and software remind us, one should not minimize the importance of taking a view on the macroeconomy, industry factors, and competitive environment in which a borrower operates – conditions matter.

At BBH, when lenders come to meetings with perfect analysis of those capacity, collateral, capital, and conditions, we often ask: How would you grade the management team? How have they worked with capital providers in tough situations in the past? Do folks whom we trust trust them?

In some ways, the character is the hardest C to evaluate – the most subjective, the most judgment-laden, and certainly the hardest to pin down. In a world where scale is everything, it can also slow down your process. Taking a day trip to an inconvenient location to break bread with the owner may prevent you from meeting your quota for the year, but that meeting is critical to build the evaluation of this “C” into your process.

From Antonio’s “merry bond” in “The Merchant of Venice” to the covenant heavy structures of a 2026 private equity transaction, the central question remains unchanged: Who is on the other side of the signature?

One of our retired Partners used to say that credit is as much about sociology and human behavior as it is about ratios. It is an observation worth taking seriously. When conditions deteriorate and collateral values evaporate, character is often the only remaining line of defense between a successful investment and a permanent loss of capital. For borrowers, too, lenders with good characters ensure a human on the other side of the table who may save a business in times of distress.

In an increasingly volatile landscape, the industry would do well to continue its rigorous analytical credit research but also remember its first principles. Data matters. Structure matters. Collateral matters. But when the cycle turns – as it always does – it is attention to character that separates the benchmark outperformers from the benchmark underperformers.

The author would like to thank Alexa Arana for her contributions to this article.