// insights
We close ~2,400 files a year. The patterns inside are public knowledge — they just live behind aggregator paywalls. We publish what we learn.
We pulled 14 years of Bank of Canada data, modeled every breakpoint, and found that 5-year variable beat 5-year fixed in 11 of 14 years. The exceptions all share one signal — and it's not the one most brokers cite.
Since the stress test launched in 2018, an estimated $230B in mortgage origination has been redirected from A-lenders to B-lenders. We map the flow and identify which lender categories grew, which shrank.
Big-bank posted renewal rates average 0.34% above broker-shopped rates. On a $500K mortgage that's $3,400 over a 5-year term. We show the math with three real comparison files.
Two-year averaging, add-back rules, gross-up factors, stated-income programs — what each lender will and won't accept on self-employment income. Plus the 6 monolines that take T2125s without a stated-income premium.
Quick visual: posted vs. broker rates, every January, 2016 to today. Pattern: broker rates lead posted rates by ~3 months on the way down, lag by ~5 months on the way up.
We modeled every break-and-refinance scenario for the 2,400 mortgages we touched last year. The break-even threshold isn't a rate spread — it's a rate spread × remaining term × penalty type. Decision tree included.
// monthly digest
Rate forecasts, regulatory changes, and the new article in the library.
