Modelling Expense Growth Over A Hold Period

Most models apply a single growth rate to the entire expense column and a slightly lower one to income. That convention is tidy and it misrepresents how costs actually behave.

Expense lines have different drivers

Payroll tracks local wages. Insurance tracks underwriting conditions in a market that spans regions. Property tax tracks assessment policy. Utilities track energy prices and consumption.

Because the drivers differ, the lines move independently and sometimes in opposite directions. A single escalator averages away exactly the variation that determines the outcome.

Grouping lines by driver rather than by accounting category is more informative. Costs that respond to the same underlying force will move together, and there are usually only a handful of forces.

Some costs are fixed and some scale

Insurance, taxes and much of administration are largely fixed with respect to occupancy. They continue whether the building is full or half empty.

Utilities in common areas, turnover costs and management fees scale with activity. A building with heavy turnover incurs costs that a stable one does not, regardless of size.

This distinction matters most in a downturn, because a property with a high fixed cost share loses net income faster than gross income falls. The operating leverage runs both ways.

The compounding effect over a long hold

A small annual difference in expense growth becomes substantial across a hold of several years, and it compounds against income growth rather than alongside it.

If expenses grow faster than rents for a sustained period, net income falls even though gross collections rise. The property looks busier and earns less.

Because value is usually derived from net income, that squeeze also affects the modelled exit. The two effects reinforce rather than offset each other.

Step changes do not follow a curve

Several major lines do not grow smoothly. Property tax jumps at reassessment, insurance can reprice sharply at renewal, and payroll moves when a position is added.

Smoothing these into an annual percentage produces a model that never shows a difficult year. Real operating histories contain difficult years, and lenders test for them.

Modelling known step changes in the year they land, and adding an allowance for unknown ones, produces a lumpier and more realistic cash flow profile.

Where the assumption should come from

The property's own expense history is the best starting point, adjusted for anything the sale changes. Several years of history reveal the trend that any single year conceals.

Where history is unavailable, comparable operating data for similar buildings under similar management is the next best source, with the caveat that management quality varies widely.

What should not happen is importing a convenient escalator because it is conventional. The number carries real weight in a long hold and deserves the same scrutiny as rent growth.