Written by David Langlieb

Like most lenders, the Philadelphia Accelerator Fund typically requires that our borrowers personally guarantee our loans. While we don’t look forward to enforcing these personal guarantees and certainly don’t like taking borrowers to court, we need to secure our loans and therefore ask our borrowers to stand behind their debt. 

When analyzing a personal guarantee, lenders like PAF will run a credit check. This is generally the last thing we do in the credit analysis process. Pulling credit typically reduces a borrower’s credit score, and we don’t want to invite that unless we expect to move ahead with the deal. While the impact of a pull on a credit score is minor—generally a reduction of around eight points— we still try to avoid it unless we’re moving at full speed towards a loan approval and closing.

Credit reports are among the most misunderstood and, in some instances, the most misanalyzed tools in the underwriting toolkit. Like tax returns, personal financial statements, reference checks, and bank statements, credit reports are there to help underwriters fill out a borrower profile. We need a basic understanding of a few things—personal income, assets, and liabilities—which we tease out from tax returns and a personal financial statement. The credit report provides a second level of detail, confirmation, and/or clarification of the information in a borrower’s personal financial statement. It also includes information on current and historical delinquencies, credit limits, bankruptcies, and the number of credit pulls by other lenders.

A credit report also assigns a score. More specifically, it assigns three scores—one from each of the three major credit bureaus—which most lenders will average into a single three-figure credit score. Trouble can start at this stage because the number tells us less than you might expect and can be deceiving if viewed out of context.  

Each credit bureau uses an algorithm to generate a score. Experian, Equifax, and TransUnion each weight the inputs somewhat differently, but as a general rule, a credit score is an amalgamation of metrics informed by:

  • Debt level and length of credit history
  • Current delinquencies (money owed that is past due)
  • Historic delinquencies (money that was past due and is now up to date)
  • Charge offs
  • Bankruptcies 
  • Number of credit pulls 


What makes an excellent credit score? A borrower with a long history of borrowing money and making payments as agreed. This is, in a general sense, logical—it means a borrower pays their debts, which is music to a lender’s ears.

That said, a credit score’s quality is only as good as the algorithm itself and the information the bureau puts through its algorithm. I’ve reviewed hundreds of credit reports during my time in nonprofit lending, and I’ve got at least three bones to pick with the credit bureaus.


Limited Credit History Can Be Deceiving 

To whom would you rather lend? Borrower A has saved her money diligently and lives in fear of debt. She rents an apartment and puts away 10% of her take-home income towards a down payment on a house. Borrower A drives a ten-year-old used car she bought with cash, and went to community college using grants and support from her parents. Every month she pays the full balance on her credit card. 

Borrower B owns a personal residence and a rental property, which carry mortgages. She purchases a new car every three or four years and typically rolls negative equity into the new car purchase, generating progressively larger car loans. Borrower B has student loan debt from a private university education and pays interest only on the balance. She makes minimum monthly payments on credit cards. Borrower B was delinquent on a mortgage twice years ago but has solid W-2 income from a full-time job and an otherwise perfect history of making payments.

To me—and any sensible credit analyst—Borrower A is far and away the better credit risk. Borrower B might be on the cusp of financial disaster, particularly if she loses her job. But Borrower B will have a higher credit score. The “lack of credit history” metric is a significant demerit in the scoring algorithms and unfairly penalizes renters and folks who are careful with their finances, like Borrower A.  


Minor Inaccuracies Can Have Major Impacts

Another significant demerit is an unpaid account, either reported as a charge-off or open delinquency. This demerit is understandable and, in most cases, correct. Credit card companies, for example, will often write off debts, even absent a bankruptcy, which justifiably generates reductions in credit scores. But in my experience, a surprising number of these demerits are in error or the result of miscommunication. There are two common examples—cable companies and medical bills. 

To flesh out the latter example—a common demerit on a credit report is a $150-$200 charge-off or delinquency reported by a cable or utility company. Sometimes a customer cancels cable service, moves to a new address, and either neglects to return the old cable box or doesn’t pay the last bill. Often the cable company may not even alert the customer about these issues. The company might not have the customer’s new address or might be so severely disorganized that the paperwork never gets to the customer. But the accounts receivable department reports the demerit to the credit bureaus, and the negative indicator typically lives on the customer’s report for seven years. A charge-off has significant credit score implications, even if it’s only a couple hundred dollars. 


Revolving Availability Metrics Punish the Self-Disciplined

This is a minor point, but one which has affected my own score and therefore engenders a lasting bitterness in my heart. One aspect of typical score-generating algorithms is a revolving availability calculation—this reflects the percentage of one’s available revolving credit (including credit cards and home equity lines of credit) classified as currently outstanding debt. For example, say you have one credit card with a $10,000 limit. If, at the time of the credit pull, you have $2,000 outstanding on that card, then you have 80% availability. 

Scoring algorithms favor a high availability ratio, which makes sense on the surface. But there’s a sinister consequence to this: these metrics reward people who apply for lots of credit cards with high limits so long as they don’t abuse them. 

As anyone with a mailing address knows, credit card companies are profligate in offering their customers new cards with progressively higher limits. A customer who keeps a low limit on one credit card might do so to impose self-discipline on using that card. Somebody with a $1,000 balance on a $2,000 limit has 50% availability, while someone with a $3,000 balance on a $10,000 limit has 70% availability. Again—who is really the better credit risk? The revolving availability metric can be deceptive. 

Given these (among other) issues, sound underwriting practice considers credit score and credit reporting as tools for analysis, but ones that must be investigated and properly contextualized. Therefore, the Philadelphia Accelerator Fund does not evaluate personal credit solely based on score and why our underwriting aims to ask the right questions and gain a comprehensive understanding of an applicant’s finances.