EIV Income Discrepancies: Thresholds, Deadlines and the Verification Hierarchy
EIV is HUD's Enterprise Income Verification system. Its use is mandatory at every annual reexamination: the PHA compares what EIV reports against what the family declared, resolves anything that does not match, and documents how it resolved it. A discrepancy is not itself a finding — failing to chase it is.
The numbers that decide a file
Substantial discrepancy
$2,400
annual gap between EIV and declared
Resolution window
60 days
from identifying the discrepancy
Document currency
120 days
age limit for third-party documents
Repayment affordability
40%
cap as a share of monthly adjusted income
Mandated use: 24 CFR 5.233. The comparison requirement at reexamination: 24 CFR 5.236(b)(3). Administrative guidance: Notice PIH 2018-18.
The verification hierarchy, strongest first
24 CFR 5.236 ranks the evidence. Using a lower level when a higher one was obtainable is the finding a file review writes up — not the discrepancy itself.
- 01EIV report / up-front income verification — HUD's EIV system as the third-party source (highest level).
- 02Up-front written third-party verification — an original or authentic source document the tenant provides (e.g. recent pay stubs, benefit letter), dated within 120 days of receipt.
- 03Written third-party verification form — verification requested directly from, and returned by, the income source.
- 04Oral third-party verification — verify with the source by phone or in person; document the date, person contacted, and information confirmed.
- 05Tenant declaration / self-certification — a signed statement from the family, used only as a last resort when no higher level is obtainable.
The period of income
EIV reports on a lag, so the window it covers is not the window you are certifying. The period runs from 15 months before the effective date to 3 months before it — see HUD Handbook 4350.3 REV-1, Exhibit 9-7. Reading a report as though it covered the certification period is how a clean family file acquires a discrepancy that was never there.
Common questions
- Is using EIV optional?
- No. 24 CFR 5.233 makes EIV use mandatory for PHAs administering the covered programmes, and 24 CFR 5.236(b)(3) requires the PHA to compare EIV data against the family's declared income at reexamination.
- What counts as a substantial discrepancy?
- A gap between EIV-reported and family-declared annual income above the threshold shown on this page. Above it the file needs a documented resolution rather than a note; below it the PHA still reviews, but the evidentiary bar is different.
- How long do we have to resolve a discrepancy?
- The resolution window shown on this page runs from when the discrepancy is identified. Notice PIH 2018-18 is the administrative guidance; the underlying comparison requirement is in 24 CFR 5.236.
- What is the verification hierarchy?
- A ranked order of evidence, strongest first, with tenant self-certification last and only where nothing higher can be obtained. This page lists all five levels as the engine applies them. Using a lower level when a higher one was available is the common file-review finding.
- What is the period of income?
- EIV reports wage and benefit data on a lag, so the window it covers is not the window you are certifying. This page shows the months before the effective date that the period spans, per HUD Handbook 4350.3 REV-1, Exhibit 9-7.
- How is a repayment agreement sized?
- Affordability is capped as a share of the family's monthly adjusted income — the percentage this tool applies is shown on this page. A repayment agreement that ignores affordability is one the family defaults on, which converts a recovered overpayment into a termination.
What this page is
A reference, not legal advice. Every figure above is read from the same module the Honestkey agency tools compute with, and each names the rule it comes from so your staff can check it. Honestkey prepares and documents determinations — it does not submit to PIC or TRACS and it is not a substitute for HUD systems or your own counsel.