How to Spot a Fake Pay Stub on a Rental Application
Forged income documents are the most common form of rental application fraud, and AI generators have erased the old visual tells. Here is what still works: the arithmetic, the metadata, the cross-document check, and verification at the source.
Why the Pay Stub Is the Weak Point
Most rental application fraud is not identity fraud. It is income fraud, and the pay stub is where it happens. An applicant who is short on income does not need to invent a person. They only need to raise a number on a two-page document that the landlord has no independent way to check.
The scale of the problem is significant by industry estimates, though the exact figure depends on who is counting and what they are counting. Snappt, a document fraud detection vendor, reported analyzing 1,462,338 applicant submissions in 2025 and identifying more than 86,000 edited applications, an average fraud rate of about 5.1 percent. Across a larger corpus the same company has put the rate of applications containing fraudulent income documentation closer to one in eight. Vendor data is not neutral, and these numbers describe multifamily portfolios that use fraud detection rather than the whole rental market, so treat them as a signal about direction rather than a precise base rate for a small landlord.
What is not in dispute is that the pay stub carries a disproportionate amount of weight in a screening decision. It drives the income-to-rent calculation, which is usually the single largest factor in an approval. A document that can be altered in a few minutes should not be the load-bearing element of that decision, and increasingly it cannot be.
What AI Document Generators Changed
The traditional advice for spotting a forged pay stub was visual. Look for mismatched fonts, a blurry employer logo, misaligned columns, the letter O used where a zero belongs, or obviously rounded numbers. That advice was written for an era when forgeries were made by hand in a word processor by someone without design skills.
That era is over. Generated documents now reproduce the layout conventions of the major payroll providers accurately, including field ordering, deduction labels, and typography. They no longer contain the visual artifacts that landlord checklists were built around. A stub that looks perfect at a glance tells you nothing, and a landlord who is still screening by eye is now getting a false sense of security rather than a real check.
This has a practical consequence worth stating plainly. Visual inspection has moved from being a primary control to being a low-yield first pass. It still catches the careless forgery. It does not catch the competent one, and the competent one is now cheap to produce.
Template Reuse Across Applicants
One pattern that has emerged is the reuse of a single fabricated template across many applications, sometimes described in the industry as template farming. Because the same base document is recycled, forgeries from the same source share structural fingerprints even when the names and numbers differ. This is invisible to a landlord reviewing one application in isolation, which is part of why detection has moved toward services that compare a document against a large corpus rather than against a checklist.
The Arithmetic Check
Layout can be copied. Internal arithmetic is harder to fake consistently, because a forger who edits the gross pay figure has to propagate that change through every dependent field. Many do not. Start by checking whether gross pay minus the itemized deductions equals the stated net pay, on that stub and on every other stub provided.
Payroll taxes give you fixed reference points. For 2026 the employee share of Social Security tax is 6.2 percent of wages up to a taxable wage base of $184,500, and Medicare is 1.45 percent with no wage cap, plus an additional 0.9 percent Medicare tax on wages above $200,000 for an individual. Those percentages are set by federal law and appear on essentially every legitimate stub for a W-2 employee. If the Social Security line on a stub is not close to 6.2 percent of the period's gross wages, or the Medicare line is not close to 1.45 percent, something is wrong. Small variances happen with pre-tax deductions and mid-year wage base crossings, so look for a meaningful mismatch rather than rounding noise.
Then check the year-to-date columns. Ask for two or three consecutive stubs and confirm that each period's gross adds to the running YTD total, and that the YTD figure is plausible given the pay frequency and the date. A stub dated in the third week of March showing a YTD gross equal to nine months of pay is a straightforward failure. Forgers who edit a single stub frequently forget that the YTD number on the next stub has to move by exactly the amount of the period between them.
Looking at the File, Not Just the Page
A pay stub submitted as a PDF carries information that is not printed on the page. PDF metadata records which application produced the file and when it was created and last modified. A stub genuinely issued by a payroll system will typically show a producer consistent with that system. A stub that reports being produced by consumer image editing or design software, or one whose modification date is days after its creation date and shortly before the application was submitted, warrants a closer look.
Metadata is suggestive rather than conclusive, and it is important not to overweight it. Applicants legitimately print stubs to PDF, scan paper copies, or download them through an employer portal that re-renders the file, and any of those produces metadata that looks nothing like the payroll provider's original. Treat an anomaly here as a reason to verify income another way, never as proof of fraud on its own.
The same caution applies to the older visual signals. Inconsistent fonts within a single field, text that sits slightly off the baseline of its row, or a logo whose resolution differs sharply from the surrounding text are all worth noticing. They remain real indicators when they appear. They simply cannot be relied on to appear.
Verify at the Source Instead of Auditing the Paper
The structural fix is to stop treating the document as the evidence and start treating it as a claim to be confirmed. Income exists in three places that an applicant does not control: the employer's records, the payroll provider's records, and the bank account the money lands in. Any one of those is stronger evidence than a stub, and forging all three is a materially harder problem than editing a PDF.
Direct employer verification remains the most accessible option for a small landlord. Call the employer using a number found independently rather than the number printed on the stub, since a fabricated stub will list a phone that reaches the applicant or an accomplice. Confirm employment status, job title, and start date. Many employers will not disclose salary as a matter of policy, but confirming that the person works there in the stated role already eliminates a large share of fabrications.
Payroll and bank-linked verification services confirm income at the source through the applicant's own authorization, and several are available to landlords through screening platforms. TenantFort's document processing runs submitted income documents through automated analysis for internal inconsistency and manipulation indicators, and its verification workflow sends employer and prior-landlord verification requests directly rather than relying on contact details supplied by the applicant. Whatever the tooling, the principle is the same. Confirm the income where it actually lives.
Cross-Checking Documents Against Each Other
Requesting two or three consecutive pay stubs plus bank statements covering the same period creates a set of documents that all have to agree. Net pay on each stub should appear as a deposit of the same amount in the bank statement, on or near the pay date, at the stated frequency. A forger who alters the stub but submits a genuine bank statement produces an inconsistency that requires no forensic tooling to see.
Watch for deposit patterns that do not match employment. Direct deposits from an employer arrive on a regular schedule in consistent amounts from a consistently named originator. Irregular transfers, round-number deposits from individuals, or a deposit stream that begins only a month or two before the application are all worth asking about. There are innocent explanations for each, including recent job changes and cash-based work, which is why the correct response is a question rather than a denial.
Consistency checks also apply across the application itself. The employer named on the stub should match the employer on the application and the one confirmed by the verification call. The pay frequency should match the dates. The address on the stub should be reconcilable with the applicant's stated history. Fabrications tend to be internally coherent within a single document and incoherent across a set.
What to Do When a Document Looks Wrong
Apply the same process to every applicant. If pay stubs are audited, bank statements requested, and employers called for one applicant, that has to happen for all of them. Selective scrutiny is a Fair Housing exposure regardless of whether the underlying suspicion turns out to be correct, and a consistent written verification procedure is the best evidence that scrutiny was not selective.
Give the applicant a chance to respond before deciding. Documents look wrong for reasons that are not fraud, including payroll corrections, retroactive raises, garnishments, and employers that use unusual stub formats. A short written request for clarification or for an additional document costs little and resolves a meaningful share of these cases.
Be careful about the compliance layer of a denial. If the decision to deny is based in whole or in part on information in a consumer report obtained from a screening company, the Fair Credit Reporting Act requires an adverse action notice to the applicant. A denial based purely on the landlord's own review of documents the applicant handed over directly is a different situation, and the safer practice when a screening report is anywhere in the file is to send the notice. Document the reason for the decision either way, and consult counsel before making a written accusation of fraud, which carries its own risks.
Frequently Asked Questions
Is submitting a fake pay stub illegal?
Fabricating a document to obtain something of value can implicate state fraud and forgery statutes, and where the document is transmitted electronically or through the mail there may be federal exposure as well. In practice, prosecution of a rental applicant is uncommon. The realistic consequence is denial of the application, which is why the landlord's priority should be catching the forgery before move-in rather than pursuing it afterward.
Can a landlord require bank statements in addition to pay stubs?
Generally yes, provided the requirement is stated in the written screening criteria and applied to every applicant. Some jurisdictions restrict what a landlord may demand or require that alternative forms of income documentation be accepted, particularly for applicants using housing vouchers or non-wage income. Check state and local rules, and make sure the criteria accommodate self-employed and benefit-income applicants who have no pay stubs at all.
What if the applicant is self-employed and has no pay stubs?
Ask for the equivalent evidence rather than penalizing the absence of a stub. Tax returns, 1099 forms, profit and loss statements, and several months of business bank statements together give a reasonable picture. Build this alternative path into the written criteria in advance so that self-employed applicants are evaluated by a defined standard rather than an ad hoc judgment.
Does calling the employer actually help if the stub is fake?
It helps considerably, but only if the phone number is obtained independently. A fabricated stub will list a number that reaches the applicant or someone assisting them, so the number printed on the document verifies nothing. Look the employer up separately, call the main line, and ask for human resources or payroll.
Can automated fraud detection replace employer verification?
It is best treated as an additional layer rather than a replacement. Automated analysis catches manipulation and internal inconsistency at a scale and speed no manual review matches, and it flags documents that look correct to the eye. It still cannot confirm that a named employer exists and employs the applicant, which is what a verification call establishes. The strongest process runs both.