ACT I
Project Granite.
The elevator jarred open. Fluorescent light. My eyes watered. The bodega coffee from three stops ago had done nothing.
I was early. Not early enough to beat the front desk assistant. She’d been here an hour already. Smile. Salutation. I barely heard her.
Project Granite.
My neck was too tight to fake more than a nod. She rose toward the kitchen. She knew what I needed.
Last night was Sam’s dinner. Prime broker event. Japanese whiskey at a sushi place, Sam’s bank picking up the tab. Rob came with me. The assistant knew. She always knew.
She was good at her job. I appreciated that.
Right turn out of the elevator bank. Badge. Door. The trading floor stretched north to south, same layout as every institutional fund everywhere.
Seven years ago, this felt like potential. Now half the floor watched for weakness. The other half ignored me until they needed something.
Project Granite.
Early hour. Floor twenty percent full. Developers in the south corner, deposited by suburban trains. A few hungry analysts who lived close. The rest would filter in within the hour.
My desk sat in the row closest to the MD offices on the north wall. Their views: 58th Street and beyond. My view: an interior wall on the left, an exterior window facing south several desks to my right.
Once that skyline meant I’d made it. Now the buildings blurred together. Glass, steel, too many hours inside them.
I dropped into my chair too hard. The shock rattled up my spine. Elbows on desk, fingers at my temples. The yellow Bloomberg keys swam in front of me. Too bright. Too much. Too early.
Twenty-eight rows of black screens. CNBC on mute, mounted every four rows. The floor was dark and quiet. It wouldn’t last.
The assistant arrived with two mugs. Tea with honey for my throat. Coffee from the stale pod grounds that lived in every office everywhere. I managed a warmer smile this time.
The bodega coffee had burned the roof of my mouth. I took a gulp anyway, ignoring the new burn. My stomach clenched. A dull ache I’d stopped questioning years ago.
Project Granite.
I hadn’t always sat this close to the MDs. When I first moved into this row, it felt like arrival. Years of choosing work over everything else. Nights, weekends, AUM over all alternatives. Friends and family would still be there. Deals wouldn’t.
They told me I was promising. Entrusted. Director of Acquisitions.
I couldn’t remember when that started feeling like a weight in my stomach.
I’d come in early to prep for my 10am with Brian, Senior MD of Originations. I’d spent the weekend building pricing models for a portfolio that landed Friday afternoon.
Friday afternoon. Sellers know exactly what they’re doing. Everyone knows.
Every deal got a name. Project plus noun.
This one was Project Granite.
Three direct reports. Two associates, one VP. All named Mike.
The Mikes knew the routine. Divvy up the deal. Stand up models. Run comps. Draft the deck. Stitch it into a memo for Investment Committee sign-off.
They were good. Trained. A few years in. I could trust them to run the traps, clean data, build the cases. They’d check in. I’d brief Brian. We’d decide where to push, where to hold back.
Then we’d lock in a Base Case. The almighty set of assumptions that generated The Price. The number the IC would use to deploy millions.
But this wasn’t our usual deal.
The portfolio’s metrics were solid. Too solid. Engineered for palatability. The structure was layered, deliberate, more complex than the usual slicing and dicing. And like always, abstracted away for the suited class on 21st floors everywhere. Far from where assets actually sat. Far from where cashflows actually happened.
The Mikes were good. But this needed a fresh approach they didn’t have yet.
So I’d spent the week rebuilding our models. More robust. More sensitive to the structure’s quirks. That meant good news and bad news for Brian. I had to be careful how I presented it. Push too hard and he’d bring in Jon to “help” with our assumptions.
Jon didn’t care about deals. He cared about dressing sharp and working relationships sharper. The rest of us knew his style would hit the P&L eventually. He’d admitted it to me once, late night, both of us still at our desks.
“You gotta take big swings. Only way to get ahead. If a deal goes bad, a smart player sees it coming and jumps out early. Let your replacement take the blame.”
His desk sat at the corner by the window. Senior Director. Best seat on the row. We reported to different MDs, but that never stopped him from finding an angle. A favor. An offer framed as opportunity that happened to benefit him most.
Conversations with Jon left me cold. I’d found a way to coexist. But I wanted him nowhere near my deals, my team.
Brad was the one to worry about. Him, I couldn’t keep out. Jon’s boss. MD. He’d elbow into every promising deal because he could.
I’d seen the pattern before. He’d claim 20% of the credit for 2% of the work. If it went bad, he had only had a minor role. Such a shame Brad wasn’t given more responsibility sooner.
Brad was the proof that you can climb high enough and stop pretending to do the work. You become insulation between decisions and consequences. Vacation with the same people. Same private schools for the kids. Pretend you earned it.
For now, neither Jon nor Brad had arrived. The floor would fill soon. Pressure would thicken. No shouting here. Just the murmur of men moving money. Occasional laughter between spreadsheets.
I had mediocre coffee, an expensive chair, and a long day ahead.
If the numbers held, the sword stayed balanced another week. Shave, shower, suit. Good enough data. Clean enough analysis. Never mind what lies beneath.
The Friday afternoon weekend-ruiner came from Jeff. Broker-dealer. Frequent counterparty. The hooks were the same ones I’d seen on every large transaction. Only so many ways to dress up moving risk from one portfolio to another.
‘Only a few of our best clients are being shown these pools on a first-look basis before we market more widely.’
Translation: We want you working fast and give us a quick bid over our reserve price.
‘No offer date set, send in best bids early for best execution.’
Translation: Feel free to overpay.
‘Innovative new pooling structure for traditional cash-flowing assets. Current tranches are structured to have additional credit enhancements above closest comparable deals sold at discounts to par this year.’
Translation: New animal. We did some math to make it look cushioned so you don’t lowball us.
‘Additional derivatives for the pools are available to help you hedge or tailor your exact risk profile for this exciting opportunity.’
Translation: Side bets dressed as risk management. Build your own bowl of contingent outcomes. Just bake in our fees.
‘Thank you for your consideration and please promptly reach out to your sales representative for data room access.’
Translation: Jeff’s catching the 4:50 train. Move your ass.
‘Broker-Dealer LLC makes no representations or warranties to the validity of the assets and buyers must exercise their own diligence. Conflicts of interest can and will occur…’
Two pages of this. You’re a big boy with a big boy job. Don’t sue us. We told you the risks. Our job is to make money from every angle, and we will.
1:47 PM
The meeting with Brian went well enough.
Five minutes of small talk. Deeper than usual. I liked Brian. Late 30s, youngest on the Investment Committee. Marathon-lean. Midwestern. He’d risen fast during the Crisis by being competent and unthreatening while his bosses worried about their own jobs.
I’d started copying his communication style. Concise. Multiple paths forward. He distilled complexity without dumbing it down.
He was one of the few people at the firm I wanted as a mentor. But he kept everyone at a polite distance. My predecessor had pushed too hard and been rotated out.
I was careful not to make the same mistake.
The meeting followed our usual rhythm. Brian had the Granite pricing exhibits on his screen. I walked him through the collateral. We both knew how small assumption changes could swing P&L across tranches.
I explained the data issues. The structure’s complexity. Why I’d rebuilt the models instead of using what we had.
He trusted me. To a point. Doing something new meant introducing unknown risk. I could see him weighing it.
The pool averaged out to average. Maybe better than average. At a price that was almost too good. Cashflow metrics, LTVs, collateral risk, all in line. Unrated, but performing. No obvious distress.
Big portfolio. Only the largest firms could look at it. Move fast and you get the sweetheart deal.
Sometimes the industry just hands you a gift. A bonus waiting to happen if you don’t fuck it up.
I almost let the Mikes run with it. They were thrilled. First time seeing a gift like this. Friday afternoon, juicy portfolio, weekend of work ahead. They were high-fiving. Fist-pumping. This is what it’s all about, baby.
I skipped the Friday corner-office session. The usual end-of-week reprieve. VPs and Directors talking markets, fantasy leagues, weekend plans. I usually stayed quiet but it made me feel human. Almost like friends.
But something in the email felt off. I needed to understand it.
The assets were scattered. Rural and urban. Varied collateral. Diversified sponsors. LTVs spread evenly across the range. No pattern in fixed versus floating debt. Sub-debt and preferred equity existed in the capital stacks but the details were thin. Performance data was spotty. Stale by months. Inconsistent.
Data problems weren’t unusual. Formatting errors. Wrong values. Sleep-deprived analysts merging sources before their MD demanded delivery. Everyone ran checks and made adjustments.
But this pool had more problems than usual. More missing values. More stale dates. More varied servicer notes. And no theme. No single asset class. No geography. No logic to how it was assembled.
That was the part that itched.
I almost didn’t call Jeff. What was the point? But the itch. So I dialed before he caught the 4:50.
He brushed off the data concerns. Didn’t even acknowledge them. That insufferable way good salespeople evaporate your questions. I was lucky to get such a diversified pool. I should bid strong and fast. Someone else already called, very excited.
I hate brokers.
I didn’t tell Brian all that. I told him what he already knew. We had models for pricing portfolio cohorts and aggregating them. Weighted average this, median that, cashflow projections by tranche. We could model underlying assets if they were the same class, then stitch all the classes together.
What we didn’t have was a way to Project Granite. Different asset types. Unusual conditions triggering unusual cashflow diversions. Unusual P&L outcomes by tranche. At least, that’s how I read it through the sterilized language, the Exhibit A formulas, the Schedule I tiering, and the 47-page definition lists.
Brian wasn’t perturbed.
“You’re overthinking it. I hear your concerns, but if we buy at your max price…” He checked his screen. Mental math. “We close out the last 1.2% of Fund 9 and start deploying from 10.”
Fund 9. Almost closed. Fund 10. Fresh capital waiting.
The firm had raised $5.6 billion for Fund IX. Fund X closed even bigger. Structured securities, whole loans, distressed portfolios. Capital needed a home. Once one fund wound down, the machine moved money from the next.
IR didn’t care what we actually did. Pension funds, sovereigns, insurance money. Billions flowed in on our track record. Not because anyone understood what a bad deal looked like.
The CRE side was stalling. Long bull run. Too much dry powder. The real estate heads still rode their post-Crisis returns, but recent vintages were thin. Clawbacks looming.
Brian heard me out. We agreed the pricing landed in the expected range. My extra work hadn’t changed the number.
It would be fine if the pool was one asset type. Fine if the data problems were normal. So why didn’t I feel fine? The whiskey from last night?
Jessica said I’d changed. More distant. Closed off. She wasn’t wrong.
I thought about the vacation we kept postponing. The gym membership I never used. The silences at dinner.
She loved the city. I used to.
We wrapped up. I stood. Stretched a shoulder. Stiff. Everything stiff. Notes in hand. Back to my desk. Back to Granite.
Brian would brief the IC at lunch later this week. His notes would help us prep for their questions.
Was he too smart to be rattled? Too secure? He’d been doing this longer. He’d worked Jeff’s side of the street and survived the Crisis.
I knew what I had to do. Not what I wanted. But there was never enough time to do what I wanted on any deal.
Always incomplete information. Always a deadline. Always a race.
The optimal price. The perfect pitch. Outbid everyone. Predict the future. Never miss. Always win. Never lose.
Fuck.
I pressed my palms into my eyes. The numbers were right. The model was fine.
So why couldn’t I stop thinking about Project Granite?