Fraud Prevention Infrastructure for Mortgage Lenders

In 2017, a file came into 360Lending that looked clean by every standard measure.

The borrower had produced government-issued identification, income documents, mortgage statements, a property tax bill, and a full appraisal. Every document was present. Every document appeared legitimate. The file was submitted to a B lender, reviewed by their underwriting team, and approved. It was only during the lender's routine phone interview — when they called the number on the borrower's credit report — that something stopped the process entirely. The person who answered the phone was the actual property owner. They had no idea any of this was happening.

What had occurred was a sophisticated fraud. The fraudster had gained access to the property under false pretenses while the real owner was absent, arranged for a legitimate appraisal to be completed, and assembled a complete documentation package using stolen identity and financial information. The appraisal was real. The appraiser had been deceived. The fraud cleared every manual verification step in a standard broker intake process, passed through a B lender's underwriting review, and reached approval stage before a single phone call — made by chance, not by design — unravelled it.

The fraud landscape in Ontario private lending

Mortgage fraud in Canada is more prevalent and more varied than most brokers and lenders discuss openly. According to Equifax Canada's Q4 2024 report, fraudulent financial documents including falsified bank statements and down payment information account for more than 90% of mortgage fraud cases. (CMP) The fraud rate declined from 0.46% in Q4 2023 to 0.19% in Q4 2024 — an encouraging trend. But Equifax Canada has explicitly warned that the decline may be temporary. A potential surge in first-time buyers and increased mortgage activity in 2025 could lead to increased fraudulent activity as consumers misrepresent their financial information to secure the best possible rates.

The 2017 case was at the sophisticated end of the spectrum. But fraud in private mortgage transactions is not always sophisticated. In another file that came through 360Lending, a borrower produced fabricated bank statements with a falsified first and last rental deposit to show that a mixed-use property had been leased at above market value. The fabrication was caught not by any automated verification tool but by a 360Lending underwriter who, on professional instinct, requested the bank statements before submitting the file to lenders specifically to confirm the occupancy status of the commercial and residential units. The fabrication was revealed when the borrower produced a poorly falsified statement containing a spelling error.

The catch was human judgment exercised at the right moment by an experienced underwriter who knew what to look for. That is not a replicable system. It is a professional competency that exists in some operations and not others, that can be present on one file and absent on the next, and that becomes progressively harder to maintain consistently as volume increases.

There is a third category that does not make headlines but is common enough to warrant naming: borrowers who hint at or ask outright whether documentation can be falsified for a mortgage they clearly do not qualify for. The request is declined. The file goes no further. But the frequency of these approaches is itself a signal about the demand side of mortgage fraud — there is a segment of borrowers who view document misrepresentation as a viable strategy, and the only consistent deterrent is a verification process they cannot game.

The infrastructure gap

FSRA's review of 101 private mortgage transactions found that missing or inadequate identity verification was present nearly 20% of the time. (Financial Services Regulatory Authority of Ontario) One in five private mortgage files reviewed by the regulator had an IDV problem. That figure reflects what is happening across the market, not just in isolated cases — and it reflects a structural reality that has nothing to do with broker competence or lender diligence.

The current verification process at most Ontario private mortgage operations follows a consistent pattern. Underwriters are trained to review bank statements and income documentation manually. Identity verification is assisted through third-party technology solutions of varying quality and integration depth. The time required varies depending on the volume of documents. There are no systematically embedded technology solutions in the intake process that flag inconsistencies, verify document authenticity, or confirm bank statement data against source records before any human review time is invested.

The gap that matters most is not the absence of any verification — it is the absence of verification built into the workflow itself. Manual review by a trained underwriter is valuable and necessary. But manual review is the last line of defense in most operations, not the first. A document that reaches an underwriter's desk has already consumed intake time, relationship capital, and in some cases lender submission time before the verification question is even asked. The 2017 case cleared every step of that process and reached lender approval. The mixed-use property case made it to underwriting with a full documentation package before a proactive underwriter's instinct caught it.

The question is not whether trained professionals can catch fraud manually. They can, sometimes. The question is whether a system that depends entirely on trained professionals catching fraud manually is adequate for a market that processed 65,233 private mortgage transactions worth $32 billion in Ontario in 2024. (Wealth Professional)

The relationship dependency problem

There is a dimension of fraud risk in private mortgage origination that is rarely discussed directly: the relationship between broker familiarity and document quality.

When brokers exhaust their existing lender relationships on a difficult file — as every broker eventually does — they begin exploring new relationships. This is not a 360Lending or Blossom Capital problem. It is an industry-wide operational reality. A broker who has worked with the same five lenders for years has an established track record with those lenders. Their submission quality is known. Their judgment is trusted. When they bring a file from an unfamiliar broker, or when a lender receives a submission from a broker they have never worked with, that established trust does not exist.

Misrepresentation of financial information — fake pay stubs, employment letters, account statements, tax slips, and false down payment information — appeared in more than 95% of applications flagged as fraudulent in Q4 2024, according to Equifax Canada. (Canadian Mortgage Trends) The documents most likely to be falsified are exactly the documents that vary most in quality between familiar and unfamiliar broker submissions. When a lender receives a file from a new broker relationship — often a difficult file that the broker's existing network has already declined — the combination of unfamiliar source and complex borrower profile creates the highest risk environment in the entire origination process.

Standardization addresses this directly. When document submission follows a consistent format with systematic verification built into the intake step, the quality of a submission is no longer a function of the broker's relationship history with the lender. A new broker relationship submitting a verified bank statement through a standardized intake process is no more risky than a familiar broker doing the same. The relationship dependency problem does not disappear — but it stops being a fraud risk vector.

The competitive advantage

A leading Canadian lending institution was spending upwards of 6 hours per application manually reviewing and compiling bank statements before implementing AI-powered document processing automation, according to a case study published by the Canadian Lenders Association citing Blanc Labs. (Canadian Lenders) Six hours per application. In a market processing tens of thousands of private mortgage transactions annually, that is an operational cost that compounds at scale.

The variance between operations in how they handle this is significant and growing. More sophisticated lenders are deploying third-party tools or building internal systems for systematic fraud detection. These solutions are often costly and technically complex to implement — out of reach for smaller operations, legacy brokerages, and individual lenders who do not have the capital or technical expertise to build them independently. The gap between what sophisticated operations can do and what smaller ones can manage is widening as fraud becomes more sophisticated.

That gap is where competitive advantage lives.

A lender who can tell brokers that every submission entering their pipeline has bank statements verified against source records before any underwriting time is invested is offering something materially different from a lender whose process depends on a trained underwriter catching a spelling error at step four. The first lender closes faster, retracts offers less frequently, and produces fewer post-commitment surprises. Brokers who know this will preferentially route their cleanest deals to lenders they trust to verify properly — because lenders who verify systematically are lenders who do not waste a broker's time with a last-minute retraction on a file that should have been caught at intake.

Mortgage Professionals Canada has consistently pressed Ottawa to prioritize a secure digital income verification system that, with borrower consent, would allow lenders and brokers to confirm income data directly from CRA records. The federal government committed in Budget 2024 and the Fall Economic Statement to begin rolling out such a system by early 2025. (Canadian Mortgage Trends) The industry's own national association and the federal government have both recognized that systematic verification is the structural answer. The question is not whether this infrastructure will exist in the Canadian mortgage market. It is which lenders and platforms will have built it into their workflow before it becomes a regulatory requirement rather than a competitive differentiator.

The fraudster who arranged an appraisal on a property he did not own, assembled a complete documentation package from stolen materials, and reached B lender approval stage before a phone call exposed him — that fraud happened in 2017, before the private mortgage market reached $32 billion, before borrower vulnerability nearly doubled, and before the volume of private mortgage transactions in Ontario reached 65,000 annually. The market has grown significantly since then. The sophistication of fraud has grown with it. The verification infrastructure at most private mortgage operations has not kept pace with either.

Ringo So is Co-Founder of Openfund, Canada's first intelligent marketplace for private mortgage origination, and Co-Founder of 360Lending and Blossom Capital.