Reading the Fed like a developer

Everyone with a Bloomberg terminal reads the Fed the same way: watch the meeting, parse the statement, count the dots. That is the right read for a trader, because a trader's position turns on 25 basis points inside a single afternoon. It is close to the wrong read for a development team, because nothing about an entitlement calendar or a construction draw schedule moves on a single afternoon. What actually moves a project's economics is a duller stack of monthly and quarterly releases that sit between meetings, and almost nobody in a development shop reads them in the order that would tell them anything.

Here is that order, and what each one is actually for.

Start with SLOOS, because it is the only series that tells you what your own lender is thinking

The Senior Loan Officer Opinion Survey is quarterly, boring to read, and the single most direct window a development team gets into bank psychology before it shows up in a term sheet. It breaks lending-standard changes out by category, including construction and land development specifically, and increasingly by bank size, which matters because large banks and regional banks are no longer moving together. The January 2026 release showed banks easing commercial real estate standards for the first time since the 2022 tightening cycle began. The April 2026 release showed that easing had already split by size: large banks kept easing, smaller banks moderately tightened construction and land-development standards specifically, and the most common reason banks gave for easing at all was competition from other lenders, not confidence in the asset class. A team that reads SLOOS before a refinancing conversation walks in knowing whether the bank across the table is in an easing mood or a defensive one. A team that skips it finds out the hard way, inside the term sheet.

Then check whether anything is actually getting built rather than merely permitted

The Census Bureau's New Residential Construction release publishes housing starts, permits, and completions monthly, and the gap between the first two numbers is itself information. Starts in May 2026 came in at a seasonally adjusted annual rate of 1,177,000, down 15.4 percent from April's 1,392,000 and down 8.7 percent from a year earlier. Permits told a calmer story: 1,413,000, down only 0.7 percent from April and essentially flat year over year. Permits reflect a decision to file paperwork. Starts reflect a decision to spend money on materials and labor in a specific month. When starts fall much faster than permits, builders are holding entitled, permitted projects rather than breaking ground on them, which is a financing or cost-of-capital signal dressed up as a housing statistic. The next release lands July 17, and the gap between those two lines is worth more than either line alone.

Industrial production's electric-power slice is a construction-adjacent series almost nobody in real estate checks

The Fed's monthly industrial production report, the G.17, gets read for what it says about manufacturing and recession risk. It has a narrower use for a developer: the utilities component, specifically the electric power generation, transmission, and distribution index, which moves with actual industrial and data-center load rather than announced capacity. It fell 0.4 percent in May 2026 even as the broader index edged up 0.1 percent. One month is noise. A sustained divergence between headline industrial production and the electric-power sub-index is the kind of early signal that shows up in interconnection queues and utility rate cases well before it shows up in a press release about a paused campus. This is not the series a trading desk watches. It is close to the only monthly federal release that speaks the same language as a power-and-permission conversation with a utility.

Read shelter separately from headline CPI, because they are answering different questions right now

This is the one that actually fooled people this year, so it is worth walking through slowly.

Headline CPI ran 4.2 percent higher year over year in May 2026, more than double the Fed's target, and it is a fair reason for a hawkish committee. But the shelter component inside that same report ran 3.4 percent, down hard from an 8.2 percent peak in March 2023, and it had actually touched 3.0 percent in November 2025, the lowest reading since August 2021, before ticking back up. Part of that uptick traces to a real-world data problem: an October 2025 government shutdown left a gap in the shelter rental sample, and the correction produced an unusually large monthly print in April 2026 once the missing data was folded back in. A developer who read the April headline shelter number without knowing about the shutdown-driven correction would have concluded housing costs were reaccelerating. They were not. The correction was catching up to a data collection gap, not describing a new trend.

Two lessons sit inside that one series. First, the inflation currently worrying the Fed is concentrated in energy and the supply shocks the Committee named directly in its own statement, not in housing costs, which changes how a development team should expect this specific tightening episode to resolve relative to a housing-driven one. Second, and more generally: a single month of any federal data series can carry a data-collection artifact that has nothing to do with the economy, and the only way to catch it is to already know the series well enough to notice when a print looks wrong before you act on it. That is the same discipline behind why we hold a brief back rather than publish a confident read on a record that is too thin or too freshly distorted to trust yet. The instinct is identical whether the record in question is a federal statistical release or a county's public comment file.

Check the market's own read last, because it moves daily and the rest of this does not

SOFR and the 10-year Treasury are the fastest-moving inputs in this whole stack, published daily, and that speed is exactly why they belong at the end of the list rather than the start. SOFR sits near 3.66 percent, the direct input to almost every floating construction loan. The 10-year closed at 4.49 percent on July 2, the input closer to how a stabilized asset gets valued. Checking these first, the way most people do, tells you what the market thinks today. Checking them last, after SLOOS, starts against permits, the electric-power slice, and shelter against headline, tells you whether today's market read is consistent with the underlying data or running ahead of it. A widening gap between the two readings, market pricing calm while the underlying releases flash stress, or the reverse, is itself a signal, and it is one you only catch by doing the slow reading first.

Reading in this order changes what you argue about in committee

Do this in sequence, SLOOS first, then starts against permits, then the electric-power slice, then shelter against headline, then the market's own daily read last, and by the time a rate decision or a dot plot lands, it explains something rather than surprising you. The Committee's June forecast, covered here in the numbers, was not a shock to anyone who had already watched shelter decelerate while headline inflation did not: that divergence is exactly why an energy-and-supply-shock-driven hawkish pivot was plausible before the meeting happened, and stayed plausible afterward.

My call on the method itself, stated so it can be checked: within two years, the gap between starts and permits becomes a standard line item development teams track quarter over quarter the way they already track cap rates, because it is currently the cheapest, most public early-warning signal for financing stress and almost nobody outside a handful of economists is charting it against their own market. If 2028 arrives and that gap is still an obscure Census line nobody in a site-selection meeting has ever mentioned, I overestimated how quickly this industry adopts a data habit that costs nothing and would have paid for itself several times over already.

Read the boring releases in order. The exciting one, the FOMC meeting, is the last thing to check, not the first.

This analysis is a source-cited research summary drawn from public records, not legal advice. It can contain errors and should be verified independently before any investment decision.