After a tenant goes bad, the fix feels obvious. Raise the bar. Ask for a higher credit score, a larger income multiple, a cleaner report. It looks like prudence, and it is easy to justify to an owner. The trouble is that it rarely does what you think it does. A higher cutoff turns away a lot of good applicants and only a few of the risky ones, because the number you raised was never measuring the thing that actually went wrong.

The Thing That Predicts a Tenancy Is The Thing You Cannot See

What makes a good tenancy is mostly invisible on an application. Whether someone pays on time when money is tight, looks after the unit, tells you early when something breaks, and stays for years: none of that is printed on the report in front of you. What is printed is a set of stand-ins. A credit score built to predict loan repayment. A stated income. A record of past events of uncertain accuracy. The gap between what you can see and what you actually want to know is an information problem, and it is one of the most studied problems in economics.

George Akerlof described it in 1970, in a paper on the used car market that later helped earn him a Nobel Prize. A buyer cannot tell a good used car from a bad one, so he pays a price based on the average, which means owners of good cars get underpaid and withdraw them, and the quality left in the market drifts downward. You can read the outline of the idea here. The mechanism is the same at a leasing desk. You cannot see quality directly, so you lean on a visible proxy and quietly start treating the proxy as if it were the quality itself. A credit score is a proxy. A proxy is not a tenant.

The Number Is Aimed At The Wrong Target, And It Is Not Always Clean

There are two problems with leaning on the score, and they compound.

The first is that it measures the wrong thing. A credit score is designed to predict how someone handles debt, not how they handle a lease, and those are different behaviors. Rental-specific history, whether they actually paid rent, how they left the last unit, whether there is a genuine eviction in their past, often predicts future tenancy more directly than a general consumer credit score, because it measures the behavior you are actually trying to understand. Someone can carry a thin file and be an excellent tenant, and someone can hold a strong score and stop paying the month their situation changes.

The second is that the proxy is often wrong on its own terms. Tenant screening reports are known to carry errors: records belonging to someone else with a similar name, eviction filings that were dismissed but never cleared, stale information that should have aged off. The Consumer Financial Protection Bureau has documented these problems across the screening industry. So raising the cutoff can mean rejecting a good applicant not only on the wrong metric, but on a metric that is factually incorrect about them.

Why a Higher Bar Quietly Costs You

Here is the part that makes tightening the filter a much weaker solution than it looks. When you raise the score cutoff, think about who actually drops out of the pool. Disproportionately, it is good applicants who carry a low number for reasons that have nothing to do with tenancy: a young renter with almost no file, someone who paid cash for years, an applicant with one old medical debt weighing down an otherwise clean history. Meanwhile the applicant who is risky in ways the score cannot see, the one whose income is about to disappear or who is hard on a unit, passes straight through, because none of that risk was ever in the number.

So a higher bar filters hard on the dimension that was already visible and does nothing about the risk that was hidden. The pool you are left with is smaller, not safer. And a smaller pool of approved applicants means longer vacancies, which in most markets cost far more than the marginal risk you were trying to design out.

Better Sreening Means More Signal, Not a Higher Bar

Akerlof's own answer to the lemons problem was not to demand higher quality. It was to close the information gap with things like warranties, certifications, and reputations, so buyers could finally tell the good from the bad. The screening version of that is to gather more real signal about an applicant, rather than raising the threshold on the one number you happen to already have.

Most of that signal is not hard to reach; it is just easy to skip in favor of the number that reads quickly. Verified income rather than a figure on a form, real rental history from a prior landlord rather than only the current one who may want the tenant gone, and rental payment behavior read in the context of the whole file, so that a 630 sitting on one old medical collection is treated as a different applicant from a 630 sitting on three recent rent-related ones. Run through the same documented criteria every time, this does something a higher cutoff never can: it narrows the gap between what you can see and what you actually need to know, so a low score prompts a closer look rather than an automatic no, and the same person gets the same decision no matter who reviewed the file.

None of this is really a technology problem, in the sense that the information usually exists. It is that the information is scattered, unverified, and easy to skip in favor of the one number that is quick to read. A structured application and screening process, and a complete picture of who you are renting to that starts before the lease is even signed rather than after, is most of what separates a confident screening decision from a hopeful one. The tool does not make the call. It makes sure the call rests on more than what was easy to see.

The Takeaway

The applicant who looks best on paper and the best tenant are not always the same person, because paper only ever shows the part that was easy to measure. Screening gets better when you widen what you can see, not when you raise the bar on the little you already saw. A higher cutoff feels like rigor, and it photographs well in a report to an owner, but most of the time it is just confidence in a proxy that was pointed at the wrong thing.

The operators who screen best are not the strictest. They are the ones who take the trouble to know the most about who they are actually letting in.

FAQ

1. Does this mean credit scores are useless for tenant screening?
No. A credit score is a useful input, especially for an applicant with little rental history, and a poor score sitting on recent rent-related problems is a real signal. The mistake is treating the score as the decision rather than as one piece of it. It was built to predict loan repayment, so it should inform a rental decision, not stand in for one.

2. Why does rental history predict a tenancy better than a credit score?
Because it measures the behavior you actually care about. How someone paid rent, treated their last unit, and whether they have a genuine prior eviction speaks directly to how they are likely to behave as your tenant. A credit score measures how they manage debt, which is related but not the same. It can look fine for someone who still stops paying rent, and look weak for someone who always pays it.

3. Isn't stricter screening simply safer?
Not usually. Raising a cutoff filters hardest on applicants who are visible and low-risk but happen to carry a low number, while the risks that are genuinely hidden pass through unaffected. You end up with a smaller applicant pool and longer vacancies without meaningfully lowering your risk. Safer screening comes from better information, not a higher threshold.

4. How do I gather more information without slowing everything down or creating fair housing problems?
Standardize it. Decide in advance what you verify, income, rental history, references, and what your criteria are, then apply that same process to every applicant. Consistency makes screening faster over time and produces better decisions. It is also an important part of fair housing compliance, provided the criteria themselves comply with the applicable laws, since the inconsistency of case-by-case judgment is a common source of both bad predictions and legal exposure.

5. What should a good screening process actually include?
Verified income rather than a stated figure, real rental history with contact to prior landlords, rental-specific payment behavior, and a careful read of any records in context rather than a pass or fail on a single score. It should also flag reports for possible errors, since screening data is frequently inaccurate, and it should document every step so decisions stay consistent and defensible.