🚨 $UPST BTIG Report Update w/ Upstart IR Comments: Upstart shares fell 8% yesterday after a BTIG report highlighted a sharp increase in UPST's 30d & 60d delinquency rates over recent months. In the charts below, reference the legend. The dark red line represents the aggregate delinquency rate, which has dragged up meaningfully. In both instances, the main source of default pressure is the 2020 securitization. This aligns closely with what was seen for the 2019 securitization, which saw peak defaults near month ~66 before coming down. BTIG argues that these late-stage loans (2019-2021) represent the "true consumer" because as loans age, early defaulters are washed out and what's left is performance mainly driven by household cash flow. Loans that suddenly default near the end of their life could represent pressure on household cash flow, or macro stress. This hypothesis is interesting, compelling and holds merit, but it is important to note that it's not the only possible explanation. The aggregate delinquency rate for Upstart securitizations (thick, dark red line) could be dragged up and mechanically inflated by high relative defaults in older securitizations. These deals may have higher defaults, but given their age, they may represent a lower proportion of Upstart's outstanding loan book. To better interpret this, we need the number of defaulted loans for each securitization, not just the delinquency rates. In theory, as older securitizations amortize, they represent a shrinking share of the overall collateral base. From a dollar-weighted perspective, this means late-life spikes in their delinquency rates shouldn’t be given the same weight as newer, larger pools when assessing today’s aggregate credit risk or funding outlook. My question for BTIG is: how exactly was the total 30-day+ delinquency rate calculated? The weighted average delinquency rate for Upstart should, in theory, improve as their newly underwritten (post 2022) loans start to become more mature versus the old loans maturing (pre 2022). Investors should find some relief in knowing that the lackluster credit performance is coming from older securitizations, which don't reflect current credit or underwriting performance. With that being said, I reached out to Upstart's IR team for a comment on these developments and here's what I heard: 1) Upstart does not endorse any third-party reports/analysis. 2) Most of the delinquency development in question is among older securitizations, which don't tell you much about current credit performance. 3) Older securitization deals likely have little outstanding (less aggregate $). 4) Absolute performance is not what matters for UPST, it's what's priced into the loans at origination (losses vs expected). 5) For Upstart's view on consumer health, refer to UMI (which is trained on Upstart's own data). In my opinion (which could be wrong), BTIG's report is factually accurate, but the interpretation of this data is more nuanced than initially believed. 1) BTIG observed similar trends across the industry (not UPST specific) 2) The securitizations in question are pre-2022, reflecting UPST's old underwriting models 3) the aggregate dollar $ value of older securitizations is likely less than more current deals, meaning the default losses would be smaller. There is definitely some cause for concern in non-prime credit (Tricolor, CarMax, 1.60 UMI). Yesterday, my main worry was that Upstart's current securitization deals were seeing wider than industry default rates, which would directly question the capabilities of their AI models and bottleneck funding. For now, that doesn't appear to be the story. What are your thoughts?