The bed that does not turn in August costs more than the one that does

The framing here is right. The assumption doing the most work in most student housing models is that occupancy is occupancy, and the non-turn bed gets credited as a revenue positive because the unit never goes dark. Your point about the second leasing calendar is the piece that almost never shows up in the expense model, and it is the more expensive problem.

A few things worth adding to your cost-center argument.

The staffing load on mid-year move-outs is disproportionate because it lands outside the concentrated August window. You are paying for a leasing coordinator to run a mini-campaign for 18 beds at a time when your comp set is fully leased and the prospective-tenant pool has already committed elsewhere. The replacement resident you find mid-year is typically a transfer or a student who failed to secure housing in August, which is its own quality-of-tenant question that lenders and property managers rarely price.

The make-ready timing is also compressed differently. A unit vacating in January does not have the August staffing depth behind it. That either stretches the make-ready timeline, costs overtime, or both. Neither outcome is in the expense ratio.

On the revenue side, your 60 percent summer rate assumption is conservative in some markets and generous in others. The number that matters is the net yield per bed after accounting for the service load you described, and I have not seen many models that produce that figure cleanly. Most roll it into a blended occupancy assumption and stop there.

The question of whether the property manager has ever tracked this as a separate cost center is the right diagnostic. If the answer is no, the NOI is overstated by an amount that will vary by property but is almost certainly not zero.

What does your current model use as the management fee structure, flat percentage of gross or a per-unit fee, and does it change between the academic-year and summer periods?

The non-turn cost framing here is genuinely useful, and the point about the management load being invisible in most underwriting is one beginners should hear early.

For anyone reading this who is new to the vocabulary: a "non-turn" bed is one where the student does not move out at the end of the academic year. They stay through summer. The "rent roll" is the spreadsheet listing every bed, what it earns, and whether it is occupied. At a glance, a bed that is never vacant looks like a win.

The problem is what that bed actually requires. A summer-occupied bed sits outside the synchronized August leasing cycle, meaning maintenance, admin, and renewal negotiation happen on its own timeline. If the property manager treats those 18 beds as ordinary occupied units rather than a separate operating population, the labor cost and the distraction from August prep get buried in the general expense line. The model looks cleaner than it is.

The underwriting error you are describing, counting non-turn occupancy as upside rather than a management cost question, is common enough that I would add one thing a beginner might not ask: ask the seller or operator for a bed-level turnover log, not just an aggregate occupancy figure. Aggregate occupancy hides which beds are turning and which are not. A property running 95 percent occupancy with 20 percent non-turn looks identical on a summary sheet to one with synchronized full turnover, until you dig into labor records.

I cannot give you a benchmark figure for what percentage is typical across the sector because it varies a lot by campus type, student population, and operator, and I would want to confirm that with someone who has managed a portfolio at scale rather than guess.

What does the rent roll on the property you are looking at actually show for summer occupancy? That number changes the whole conversation.

My two Georgia notes are the opposite problem but the same lesson, the cost you cannot see in the numbers is always the one eating the margin. What you said about underwriters treating non-turn beds as revenue upside because they never go vacant, that is a real pattern and it drives me crazy every time I see it in a deck.

The part I would push on is whether the 60 percent summer rate assumption is even consistent across markets. University towns with strong graduate enrollment or medical school populations sometimes hold closer to 75 or 80 percent of academic-year rent through summer because that subgroup signs longer leases and does not follow the August machine at all. Lubbock looks nothing like Gainesville on this. So the drag calculation depends heavily on who those 18 beds actually are.

What I have never seen anyone model cleanly is the staffing distortion. When your leasing team is babysitting mid-year vacates from the non-turn pool in October and February, those are the same weeks they should be building the waitlist for next August. That attention cost is real and it does not show up anywhere in a T12.

Have you ever seen a property where the manager actually ran the non-turn group as a tracked cost center, or is it always just folded into general operations?

The cost that gets missed is the staffing delta, not just the make-ready timing. A synchronized August turn concentrates every hour of leasing coordinator time, maintenance labor, and vendor scheduling into roughly six weeks, which is expensive but foreseeable and therefore biddable. The non-turn population spreads that demand across the full twelve months at unpredictable intervals, which means the property either carries excess staff to absorb the irregular work or the manager pulls time from the 102-bed August machine to handle the mid-year vacate. Neither shows up as a separate line because most student housing operators run a single maintenance and leasing cost pool. The distortion compounds when the non-turn beds are clustered in a specific unit type, say two-bedrooms where one roommate leaves and one stays, because now a partial unit vacancy sits on the rent roll at partial revenue while a full-unit make-ready is structurally impossible until the second occupant also clears. That is where the revenue-upside argument the original post identifies is most dangerous, because the occupied bed suppresses the yield on the vacant one without appearing in a standard occupancy metric.

The number I would want before anything else is the average days-to-re-lease on the mid-year vacates, broken out separately from the August turn. If mid-year beds take 45 days to re-lease and August beds take seven, the revenue loss on those 18 beds is not 90 days of the 40-percent rate reduction, it is potentially 120 to 135 days per bed per cycle, and the model changes materially. Do you have that figure, or has the operator only been tracking aggregate occupancy across the full year?

The assumption doing the most work in this framing is that the non-turn beds are a fixed 15 percent, because in markets where graduate students or international students make up a meaningful share of the bed count, that figure can climb to 25 or 30 percent and the two-calendar problem compounds faster than the revenue haircut suggests. The mid-year vacate is the sharper cost, not the summer rate reduction, because a bed that turns in January in a college town often sits longer than one that turns in August, and the make-ready hits when the leasing team is already stretched thin on renewals. What the underwriting models I have reviewed consistently miss is that the 102-bed August machine is priced as though it runs on full management bandwidth, but if the non-turn population generates even two maintenance calls per bed per month through summer, you have absorbed a meaningful portion of that bandwidth before August arrives. The expense ratio question is the right one to press, and the place to start is whether the property management agreement distinguishes between routine turn cost and mid-year vacancy cost at all, because most don't, which means the P and L buries the true non-turn drag in a blended maintenance line.

What is the student composition at this property, specifically the share of graduate or visa-status tenants, because that is usually what determines whether 15 percent is the ceiling or the floor?

The common misconception is that summer retention always signals strong demand, when in most purpose-built student housing markets it signals the opposite: a resident who could not or did not engage with the August leasing cycle, often because their housing situation is irregular enough that synchronized lease-up does not fit them.

The labor cost question is real, but the deeper issue is what the non-turn population does to the renewal pipeline for those specific beds. A bed held through summer on a reduced rate frequently turns over in November or February, which puts it back on the market when the primary leasing season for the following fall has already closed. That bed then either sits vacant for months or gets filled by a resident who arrived late, and late arrivals correlate with higher mid-year attrition. The summer retention rate is not one problem, it is the first event in a sequence that compounds through the following academic year.

On the expense side, the make-ready cost for a mid-year vacancy runs materially higher than an August turn because you cannot batch it with the other 102 beds. Vendors charge a premium for individual unit work outside the summer window, and the internal staffing cost per unit is higher when you cannot run a coordinated turn team.

What does the lease expiration schedule look like on those 18 beds specifically, and are they on month-to-month arrangements or short fixed terms that land outside August?