DX: Q3 2023 Update by Scott Kennedy

CWMF’s Note: Scott’s updates will eventually come out under his name. For the moment, I’m sharing them. You may notice overlapping sections with the recent update for ORC. Scott absolutely nailed this projection. BV within 1%. Earnings are much closer to Scott’s estimates than consensus estimates. The value of expertise is demonstrated again.

Scott Kennedy prepared the following article.

• This 4th earnings assessment article reviews DX’s BV and core earnings performance during Q3 2023. This article is mainly for subscribers who do not utilize chat.

• This article also discusses how DX’s quarterly change in BV and core earnings “matched up” to expectations. Earnings remain a key driver to stock performance.

• DX’s BV matched my/our expectations while its core earnings was a very minor-minor outperformance.

• No change in DX’s percentage recommendation ranges or risk rating.  DX is currently deemed appropriately valued (HOLD).

• As anticipated, a notable decrease in DX's BV for Q3 2023 which has continued during October 2023.  Core earnings stabilized but remains negative/suppressed.  I would remain cautious in this sub-sector for now.

DX’s BV and Core Earnings Q3 2023 Performance (Projected Versus Actual Results):

On 10/23/2023, Dynex Capital Inc. (DX) reported the company’s earnings results for the third quarter of 2023. Table 1 below provides DX’s BV and earnings summary.

Table 1 – DX Q3 2023 BV and Earnings Summary

Source: Taken Directly from the REIT Forum’s © Analytical Spreadsheets/Data

I was able to review DX's Q3 2023 earnings results. DX reported a BV as of 9/30/2023 of $12.25 per common share (13.7% decrease) versus my prior projection of $12.35 per common share (13.0% decrease). I consider this nearly an exact match (less than a 1.0% variance) and was well within my $11.90 - $12.80 per common share range.

I would just point out the extreme volatility, especially for the agency mREIT model, during the third quarter of 2023. BVs notably decreased during September 2023 (some peers in excess of 12%). In my personal opinion, any quarterly BV variance at or within 4% this particular quarter should really be considered an accurate estimate (larger “cone” per se). So, only a (0.7%) quarterly BV variance should be considered extremely accurate. In addition, as a reminder, the current subscriber spreadsheets are based on CURRENT BVs. For example, currently each company’s BV estimate as of 10/20/2023. In this chat note, I am comparing 9/30/2023 BV projections versus actual results. Something to mention for newer subscribers. Even though this was an extremely accurate BV projection, let me provide a full reconciliation.

First, unlike the prior quarter, DX was less aggressive decreasing the company’s off-balance sheet net long “to-be-announced” (“TBA”) mortgage-backed securities (“MBS”) position during the third quarter of 2023. This basically matched expectations. DX decreased the company’s net long TBA MBS position by ($0.5) billion or (23%) as of 9/30/2023 when compared to 6/30/2023 (based on notional value). This compared to a decrease of (42%) during the prior quarter. DX reported a TBA MBS net valuation loss of ($70) million during the third quarter of 2023. In comparison, I projected a TBA MBS net valuation loss of ($65) million. When calculated, this directly led to a BV underperformance of only ($5) million or ($0.09) per common share.