Welcome back, AI prodigies!
In today’s Sunday Special:
📜The Prelude
🏁Is Fairness About Process or Outcome?
🎯What Goes Into Calculating Your Credit Score?
🔑Key Takeaway
Read Time: 7 minutes
🎓Key Terms
Machine Learning (ML): Leverages data to recognize patterns and make predictions without being explicitly programmed to do so.
Annual Percentage Rate (APR): The total yearly cost of borrowing money, including additional fees like closing costs. This percentage provides a more accurate picture of a loan’s true cost.
🩺 PULSE CHECK
Does it feel unfair when you’re denied access to credit?
📜THE PRELUDE
Imagine you’re sitting at the kitchen counter, the house silent as midnight hits. Your face is washed in the screen’s glow, fingers tapping restlessly across the keys. A sudden buzz tickles your pocket. It’s a Gmail alert: “Your Mortgage Application Status Update! {#84386457}.”
You mentally review your financial checklist: a steady paycheck, a solid credit score, and a well-balanced budget. You nervously tap the Gmail alert to view the details: “We regret to inform you....”
Within 60 days, you’ll receive an explanation from the lender for the denial. But right now, it doesn’t feel real, it doesn’t feel right, it doesn’t feel rational. By any sensible standard, your financial checklist is solid. Despite this, your American Dream of homeownership feels farther out of reach than ever before.
Today, a lender’s evaluation of a borrower’s creditworthiness is largely determined by AI-driven algorithms rooted in mathematical precision. While credit decisions should feel objective, they often don’t. As a result, we’re constantly left wondering: “Can AI-powered credit decisions ever feel fair?”
🏁IS FAIRNESS ABOUT PROCESS OR OUTCOME?
⦿ 1️⃣ Fairness Based on Process?
In 1971, American political philosopher John Rawls published “A Theory of Justice,” where he introduced the concept of the “Veil of Ignorance.” Imagine you’re designing society’s rules, but you don’t know anything about yourself, including your race, gender, abilities, and wealth. Since you can’t foresee your status in society, you’re forced to formulate impartial decisions based on fairness rather than self-interest.
He believed justice revolved around how decisions are made, not on whether the outcomes of those decisions feel fair. In other words, a “fair society” is endorsed by rational individuals who might end up anywhere within it. The civic rules must be publicly justifiable and consistently applied.
⦿ 2️⃣ Fairness Based on Outcome?
In 2009, Indian economic philosopher Amartya Sen published “The Idea of Justice,” accusing Rawls of developing an ideal theory of justice while neglecting how it actually shapes the lives of everyday people.
He believes justice focuses on people’s abilities to do the things they value in life. In other words, justice is about expanding people’s capabilities. In this context, capabilities are essentially the freedoms people have to achieve the outcomes they value in life, whether it’s fame, wealth, or education.
⦿ 3️⃣ Are Credit Scores Fair?
In 1989, American data analytics company Fair Isaac Corporation (FICO) developed the FICO Score: a type of credit rating ranging from 300 to 850, where higher scores indicate better creditworthiness. It’s essentially a three-digit number based on your credit reports that helps lenders determine how likely you are to repay a loan.
According to Rawls, the credit score is justified because it doesn’t ask why a payment was late or whether debt avoidance was prudent. It only examines whether past payment behavior correlates with the ability to repay future debts.
The FICO Score is widely considered a fair measure of credit risk by objectively predicting the likelihood of repayment based on things like payment history. Despite this, many argue that credit scores perpetuate systemic economic disparities by rewarding established credit history. As a result, lower-income individuals who lack access to credit-building opportunities struggle to secure favorable financing options.
🎯WHAT GOES INTO CALCULATING YOUR CREDIT SCORE?
⦿ 4️⃣ What’s in Your Credit Score?
Although the exact mathematical formula remains a trade secret, FICO disclosed that your credit score is derived from five broad categories of credit data:
1. 🔴 Payment History (35%): Whether you paid your bills on time.
2. 🟠 Amounts Owed (30%): How much debt you carry in total.
3. 🟡 Length of Credit History (15%): The average age across all credit accounts.
4. 🟢 Credit Mix (10%): The diversity of credit types, like credit cards and student loans.
5. 🔵 New Credit (10%): Any recent credit applications and newly opened credit accounts.
🚨 Credit Profile → Risk Signal:
The input factors common among default borrowers get positive weights, and the input factors common among non-default borrowers get negative weights.
These learned weights are applied to a new credit applicant’s profile to generate a credit risk score, where negative indicates safer than average, zero indicates average, and positive indicates riskier than average.
📌 Risk Signal → Credit Score:
FICO leverages AI-powered algorithms to examine the historical data of how past borrowers with similar credit risk scores ended up defaulting on their loans. Let’s say that, historically, borrowers with a credit risk score of +1.5 defaulted about 7% of the time. That means the default probability for that borrower equates to roughly 7%.
✂️ Credit Score → Lender Decision:
🔑KEY TAKEAWAY
AI-powered credit scoring is designed to be standardized and systematic. Yet, the outcomes often carry deeply personal consequences. When something as crucial as homeownership hangs in the balance, fairness feels like it should go beyond AI-driven algorithms rooted in mathematical precision. Ideally, the human side of fairness should matter just as much as the algorithmic side.
📒FINAL NOTE
FEEDBACK
How would you rate today’s email?
❤️TAIP Review of The Week
“I worked in the electric utility industry for 12 years, including 8 years at a generating site with 4 coal plants and 1 nuclear plant. I enjoy your explanations of the new energy production methods.”
REFER & EARN
🎉Your Friends Learn, You Earn!
{{rp_personalized_text}}
Share your unique referral link: {{rp_refer_url}}
