Outside DFW TX. 15yr veteran of PE + CRE. Partner at a multifamily GP. Revitalizing an old farm outside day job.
Texas
Joined May 2025
- Tweets3.4K
- Following848
- Followers2.1K
- Likes9.2K
Pinned Tweet
I just got @thematthew as my new handle
Guess I'm no longer an anonymous paintbrush
This is who runs this account:
> Texas born and bred
> Spent my career in structured credit, PE and CRE
> Opened up my own shop 5 years ago
> Love my family and love where I live
> More info to follow on my company, my sidequests, and everything else I love to share
“A decade of pain” may be in the cards.
Just for the Texas big 4 markets, we’re still YEARS away from simply clearing excess vacancy from the metros.
In order of fastest to clear to slowest:
1) Austin
2) Houston
3) DFW
4) San Antonio
Dinner with an exec/friend at a large national property management company tonight. This person oversees ~150k apartments; access to an insane amount of data.
Current multifamily market in most of the country is worse than S&L, .com bust, or the GFC. Worst they’ve seen.
Class A mostly fine, decent rent growth. However:
- credit scores at class A have gone down meaningfully over the past 3 years.
- more people are paying rent via credit card.
- average bank account balance noticeably lower than even 2 years ago
- class A tenants stretched thin, living beyond their means. 20k comes in, 20k goes out every month.
- starting to see these renters move down a tier
Class B:
- simply too many apartments, we built too much new stuff. Ripple effect is enormous.
- fewer renters tomorrow than today. Same store foot traffic noticeably down year over year. Some markets down 40-50%
- considering dropping rents below $700 a month on some 1beds in major southern markets.
Class C:
- looking at degrading NOI. Rents aren’t growing for a long time, but expenses keep creeping up
- more renters are one missed paycheck away from disaster than they can remember seeing in decades
- tenants aren’t paying rent because they simply don’t have it
- wage growth has not kept up
Lenders:
- some major investment funds, household names, haven’t paid their mortgages in 2-3 years. Banks won’t foreclose because “X is a big client.” Mexican standoff.
- a roof leak gets ignored, becomes water damage, then mold, then a down building. Buildings are rapidly deteriorating as some owners choose cash preservation over funding losses.
- number of pre-foreclosure and REO deals is the highest as a % of total units they can recall.
- data providers know it’s bad, but they’re paid to sell optimism to owners/lenders. So every forecast is “it will probably get a little better next year maybe” and that gets regurgitated at investment committees all over.
I asked “how do you think this ends?”
“Not sure. My gut says slowly then suddenly, and things will get really bad. Been doing this 40 years, you can feel it in the air. But, there’s a chance it’s just a decade of tolerable pain. I’m praying for that.”
Submarket analysis for DFW multifamily below:
Image 1 = snapshot of concessions in submarkets on new builds 2023+
Image 2 = years to clear excess supply based on (a) recent renter demand stats for most, and (b) for exurbs it uses NCTCOG forecasts
Note #1 = NCTCOG's forecasts are used for exurbs because housing statistics where there was very recently just corn fields and cattle need adjustments
Note #2 = submarkets in teal are declining in renter households
Note #3 = Richardson is odd to see, because it's ostensibly a good school system... but occupied units were 21,300 in 2017 and there are 21,200 now but with 2,500 vacant units to fill (so renter demand is essentially flat over ~10 years but +10% units appeared recently)
Note #4 = Mesquite is a similar story to Richardson, but this submarket also includes Forney so locals know the growth pace on that southeastern fringe
I tried this myself and the first song I made is so bad that I'm afraid Sturgill Simpson is going to murder me, violently
AI generated music is 1000x better than real music.
Today I canceled Spotify and SiriusXM because I finally tested Suno.
What sold me was the first prompt I gave it:
“My favorite artists are A Day to Remember, Bring Me The Horizon, Spiritbox, Bad Omens, and Sleep Token.
Analyze all their hits and generate a song I’m likely to love based on those artists.”
It one shots an absolute banger in 10 seconds.
I'll link the song in the next tweet for anyone curious.
Why would I ever go back to Spotify/SiriusXM when I can generate music perfectly tailored to my taste on command, for free?
Not affiliated with Suno in any way btw, it was just one of those moments where I was so impressed with AI’s capability I had to say something about it.
This was my style prompt:
"I like old cowboy yodeling songs where they used to sing to cattle, but I also metal bands if they're melodic like Tool or Sabaton, but also do occasionally listen to chill electronic music like Tycho and Rufus De Sol, while also getting into weirder electronic music like Aphex Twin, though I will come back regularly to rock/indie like Radiohead because I'm a millennial, but when I'm driving I like to listen to good ol' outlaw country and Texas country music from 1970s-1990s"
Rational actors will see that a majority of population voted to take some % of their wealth
They would rationally know voting pop don't change drastically every year
They can rationally assume that same majority could put it on the ballot to take some % of their wealth again?
Let’s be clear: this is a highly inappropriate way to model the effect of Prop 40 on taxpayer behavior.
For one, it’s a basic violation of rational expectations, a cornerstone of modern economics.
In Jaros and @joshrauh world, the billionaires act irrationally: they respond to a policy that does not exist and will never exist.
Economics students around the world are taught this is not an appropriate way to model behavior — but apparently in this context (fighting Prop 40) it’s OK.
Multifamily and single family are tied at the hip.
If you're in multi but aren't looking at both and understanding consumer/household choices - you're ignoring half the field.
And if your metros are permitting housing faster than populations are growing... anyone's demand and absorption projections, as good as they seem, don't make you money.
For months I was skeptical of using linear with claude code, codex, etc.
I've avoided it for the same principle I've avoided obsidian and all the other middleman apps bandied about:
> they often get in the way of actual software development by needing an additional layer of context
> they often need ongoing maintenance to keep it accurate
> you can build something 80% as good that you can maintain yourself easily and not have a company involved in your business
> inherent skepticism of future changes in features and pricing that usually accompany these marketing pushes based on a career of seeing exactly this same playbook play out going back to the "Big Data" era and earlier
But actually... linear does connect seamlessly with every agent harness I've used so far. And more importantly, feature development and bug tracking is followed more closely and more accurately than my markdown files and it self-updates itself beautifully so far.
Not a paid endorsement, don't know them - just sharing my last few weeks of experience using linear in my codebases. Github + linear is a good combo.
As I do get to productionizing and publishing apps for wide use, I've heard good things about sentry for catching bugs and issues in real time in actual user usage... TBD on my experiences there. Methodically testing one system at a time here.
You're a Texas broker and you're asking me to buy B/C class multi in B/C pockets at a 6.0% - 6.25% cap rate?
And claiming that's "market"?
And there's a 5.7% cap rate in supply-constrained areas near OpenAI?
Where rent growth > inflation there?
Only 30bps difference?
TO: all Pension CIOs
FROM: me (Matthew)
DATE: October 20, 2026
SUBJECT: Fees and Expenses
I exclusively drive beaters.
Please DM me for sub docs and wiring instructions for our fund.
Sincerely,
Matthew
Most leaders are letting AI write their LinkedIn posts.
That’s a mistake.
Not because the technology lacks capability.
Because leadership requires something deeper.
Yesterday, between reviewing quarterly priorities and asking Claude to summarize a document I had already summarized with Claude, I paused.
Something felt off.
The insights were actionable.
The tensions were nuanced.
The tapestry was deeply human.
And yet.
I couldn’t shake the feeling that we were losing something essential: the courage to have a thought before formatting it.
Three tensions emerged:
→ The authenticity paradox. We’ve never had more tools to amplify our voices. We’ve never sounded more like the same regional vice president.
→ The frictionless mirage. We’re turning hard-won operational scar tissue into “a powerful reminder.” Sometimes the reminder is that a vendor didn’t show up
→ The curated surrender. We’re outsourcing our most personal convictions to a model that just helped someone else write the opposite conviction with equal depth.
I’ve been sitting with this.
Not solving.
Not optimizing.
Just sitting.
Because true leadership isn’t about outsourcing linguistic patterns that resonate.
It’s about opening a blank document, confronting the complexity of your own experience, and asking Sonnet to make it sound less AI-generated.
The best leaders aren’t just prompting for engagement.
They’re holding space for their own baseline brain function.
How is your team leaning into bespoke cognitive friction this quarter?
♻️ Repost to help someone reclaim their authentic voice using these exact words.
Does this make the rooftop Koreans from the LA riots also “right wing”?
Didn’t realize defending the place you were born (or even just call home) from destruction and crime could be described as something on the political or ideological spectrum
It’s just sensible practice
Sunday scenes from the farm:
Temperature dropped overnight, pond is steamy on the morning walk and looking cinematic.
Cultivate land, steward assets, pass it on.
It can’t be both
“Anthropic is going to eat every other company and take over the world and all our jobs
AND
“Your ability to repay is priced as if you are a existentially in doubt”
In my head the team used analog transistors and vacuum tubes purely for the laughs and Apollo mission throwback vibes
Hate to dox myself a bit but have worked on just about every fighter jet the US has. I have seen some impressive technology over my career but this is a whole other level. I have so many questions and need to know how this works. How much edge compute is needed to make this happen? My first thought is dual processor + 1TB ram. That way it can handle the fuckquintillion process per second. There was zero correction it nailed it 100%.
