The rent number on a rent roll gets all the attention. The lease structure next to it quietly decides who pays for the future, and over a ten-year hold, that decision is frequently worth more than the rent spread that distracted everyone at acquisition.
The mechanics are familiar to anyone operating multi-tenant retail. Under triple net, expense growth is the tenant's problem: taxes reassess, insurance spikes, CAM inflates, and the landlord's net position holds. Under modified gross, some or all of that growth lands on the landlord, and where a base-year stop exists, the protection erodes in a specific way: the stop is frozen at a point in time while expenses are not, so the landlord absorbs the growth above it, compounding annually, silently, on a line item nobody reprices until renewal.
This is why a modified gross lease at a higher face rent can be a worse lease than an NNN lease at a lower one, and why the comparison cannot be made in your head. It depends on the expense load, the growth rate, and the years remaining. At low expense growth the MG premium may genuinely be worth more. Run the same lease at heavier growth and the premium is gone within a few years, after which the landlord is paying the tenant's inflation out of the deal's cash flow.
Conversions are where this becomes an underwriting event rather than a background condition. A lease that converts from modified gross to NNN in year three does not just tick NOI up in year three. It changes the slope of NOI for every year after, because it transfers the compounding to the other side of the table. Value the conversion as a one-year bump and you have underpriced it; miss it entirely, which blended-average models do by construction, and you have mispriced the building. The reverse conversion, NNN to gross, is a concession sometimes traded for term or credit, and it deserves the same slope-not-step treatment as a cost.
The widget below runs one lease both ways. Same suite, same face rent, one modeled as NNN and one as modified gross with a base-year stop, across an expense load, a growth rate, and a hold you choose. The output is the cumulative NOI difference, which is usually the moment the structure stops being fine print.
A real building holds both structures at once, plus escalations, stops set in different base years, and conversions negotiated tenant by tenant. That heterogeneity is exactly what per-lease modeling exists for. PropCalc™ carries each lease's actual structure through the full pro forma, so the building's NOI slope is the sum of real leases rather than an average that never signed anything.