Skip to main content


Clouded judgment: mood influences credit approval decisions and outcomes

Ran Duchin

Ran Duchin

You are more likely to get a loan approved on a sunny day—especially an unexpectedly sunny day.

So concludes a new study on the role of sentiment in managerial decision making that is co-authored by Ran Duchin, an associate professor of finance at the University of Washington Foster School of Business.

Duchin’s analysis of the $1 trillion American mortgage market reveals that the mood-boosting power of an unseasonably sunny day increases the approval rate of marginal loan applications, while the mood-busting quality of unusual cloud cover diminishes the approval rate.

Moreover, this climatologically-driven sentiment produces a measurable economic impact. Specifically, loans approved in the throes of unexpected sunshine are more likely to wind up in default.

“By comparing the variation in local sunshine to the fate of mortgage applications nationwide, we were able to observe the effect of sentiment on each loan decision,” says Duchin, the William A. Fowler Endowed Professor at Foster. “Even better, we observed the economic outcome that results from sentiment. And since it’s such a huge market, the numbers we arrive at are pretty significant.”

Living laboratory

Examining the effects of mood on behavior is a difficult proposition outside of a laboratory.

But Duchin and his co-authors—Kristle Cortés of the Federal Reserve Bank of Cleveland and Denis Sosyura of the University of Michigan—discovered parallel sets of real-world data that were precise, plentiful and nearly perfect for the task. Namely, the consumer home loan applications and hourly weather conditions recorded in every county across the US from 1998 to 2010.

To isolate the effect of mood, the researchers focused on days of unseasonable sun that can deliver an emotional charge and days of surprising cloud cover that can dampen one’s outlook—the rare clear winter day in Seattle, for instance, or the odd overcast summer day in San Diego.

They also threw out loan applications that would obviously be accepted or rejected in any weather (or any emotional state)—the applicant with 50 percent down payment and an 810 FICO score, say, or the one with spotty employment and a 570 FICO. What remains are the applications at the margin: judgment calls.

Forming those judgments are professionals, “people who should be making their decisions in a well-informed, objective way,” Duchin says. “And, in fact, loan officers have a well-defined set of parameters that should be driving their decisions. So it’s easy for us to see what they should have done, what they did on average, and how they deviate on days when it’s surprisingly sunny or cloudy.”

Economic impact

What does weather-induced sentiment do for marginal loan approvals? Duchin, Cortés and Sosyura calculated a significant impact. A rough estimate of the extra credit approved on one perfectly sunny day relative to one fully overcast day is about $150 million nationwide, or about $91,000 per county.

This effect is asymmetric. Approval rates increase by 0.8 percent on sunny days. On cloudy days they decrease by 1.4 percent.

Of course, the laws of economics allow that even marginal mortgages approved by sun-blinded loan officers should be acceptable so long as those loans carry a higher rate of interest. “As with everything in finance,” Duchin says, “if it’s priced correctly, then who cares? As long as you charge a higher interest rate to offset the extra risk you take on marginal loans, then you’re fine.”

But he found no evidence that interest rates on iffy loans approved on sunny days were adjusted to compensate for the increased risk.

Their applicants received, in effect, a gift provided by nature and, well, human nature.

There is a real downside to this gift, though. Loans approved on sunny days were shown to experience significantly higher defaults which exact a real toll on the economy. “Sentiment,” Duchin says, “has a significant economic effect.”

Discretionary lending

Given the extraordinary detail of their data, Duchin and his colleagues also were able to determine why weather-induced sentiment influences credit approval decisions, and the conditions in which the effect is strongest.

Comparing loan-to-value ratios and credit scores of approved applications, they found evidence that loan officers in a good mood are less averse to risk.

Examining credit rejection reports, they also concluded that loan officers exhibit more favorable judgment on sunny days. They don’t scrutinize as closely or critically.

Discretion matters, too. The effect of weather-induced sentiment on loan approvals appears to be greater in situations where loan officers are allowed to exercise greater discretionary judgment. This includes smaller or local lending institutions and during periods such as the housing boom when loan officers were granted unprecedented latitude on approvals.

Broader effect

Duchin, Cortés and Sosyura may have found an ideal lens through which to observe the effect of fluctuating moods on the professional decision making process. But their findings should not be limited to the consumer loan market. Human judgment plays a critical role in everything from economic forecasting to air traffic control to college admissions.

“It’s easy to think of other instances of decision-making in which sentiment might play a key role,” Duchin says. “The majority of organizational decisions rely on the judgment of lower-level officers. While we focus on repeated, well-understood decisions of trained financial intermediaries, sentiment could also influence many other agents who face ambiguity and have significant discretion.”

Duchin also points out that this analysis merely scratches the surface on sentiment. If something as slight as a change in the weather can have such a measurable impact on a person’s professional judgment, imagine the effect of more intensely personal regulators of mood—deaths, divorces and family emergencies on the downside, and birthdays, awards and love affairs on the upside.

Automation and disincentives

So, what is an organization to do to compensate for its human frailties of judgment?

Duchin says that one possible solution is to remove humans from the decision, at least partially. In the consumer mortgage industry, for instance, firms could automate at least a first level of screening. This would enable a more objective assessment of an applicant’s fiscal fitness for a loan.

Lenders also might consider imposing financial penalties on officers who approve loans destined for default. Requiring some “skin in the game” could act as a strong disincentive to faulty judgment in any weather or emotional state.

Even if people continue to be entrusted with credit approval decisions, Duchin adds that lending institutions could at least automate the pricing of risk as a fair insurance against the possibility of default.

And what about consumers? Do the study’s findings suggest that it might be wise to wait until an unseasonably sunny day to apply for a loan?

It doesn’t quite work that way. “You have to realize that the day you go in and apply for a mortgage is not the day that the loan officer reviews your application,” Duchin says. “It’s not so easy to manipulate the process.”

Clouded Judgment: The Role of Sentiment in Credit Origination” is forthcoming in the Journal of Financial Economics.