Budgeting at the community level instead of the unit-type level

This mismodeling is often driven by the need to simplify models in Excel. Not that you can’t budget at the unit type level in Excel…it’s just a LOT more complex and thus harder. We’ve thus seen some of the larger operators in our industry simplify their Excel-based models to budget at the community level leading to at least three key challenges:

1. The business doesn’t behave at the community level; it behaves that the bedroom count level.
   Three-bedroom homes lease VERY differently than one-bedroom and studio homes. In some markets, two-bedroom homes lease somewhat similarly to one-bedrooms, and in others, they do not. Whatever the case, it is difficult to model key drivers such as seasonality (see below for more on this driver), renewal rates, and vacant days between leases when you’re blending these stats instead of modeling drivers directly at the unit type level.
2. Rent-constrained unit types.
   The classic example here is “bond units.” That’s when (typically) 20% of the community must be set aside for what can also be referred to as “below market units” (BMR) for residents or within a specific range of average median income (AMI) for the area. Analogous to the community vs unit type level discussion, the same challenge of picking a single set of drivers to model a blend of units that behave very differently often causes mistakes that drive unnecessary budget variances. It takes a good bit of offline math to understand how 80% of the units moving with the market and 20% with very low rent growth behave (not to mention differences in renewal rates, delinquency, etc.)
3. Expiration mix.
   And then there’s the expiration mix. If 10 one-bedroom homes and 5 two-bedroom homes are expiring one month and then the reverse the next month, community-level calculations will miss these potentially material differences. Even if the same rent growth and seasonality applies (which often does not), the raw dollar change is higher on the more expensive homes. This expiration mix is equally an issue with differences between BMR and market-rate homes.
   
   

Occupancy as an input instead of an output

As with community-level modeling, this is often an intentional simplification to make it easier to build Excel models. But just because it’s easier doesn’t make it correct. Building a true expected value (EV) model taking into account month-to-month (MTM) leases and the possibility of multiple leases per year due to short lease terms and/or early terminations is complex, and that’s exactly why simpler models will usually introduce unnecessary variance!

Not modeling seasonality correctly (or at all)

We’ve seen models that don’t take into account demand (occupancy) seasonality, rent seasonality, or both. This will obviously cause variances, especially in those markets that are highly seasonal like the Northeast and Northwest. Another classic mistake is to have seasonality but be unrealistic about it. If your budgets show occupancy and rent both growing in the fourth quarter, then you are almost certainly making this mistake. With very few exceptions, we’re lucky to hold onto occupancy with slight rent reductions in Q4.

Bonus #4: Missing the renewal strategy

For those of you using revenue management systems or strategies that involve “capture rates” (the percent of the gap between new and expiring rent to “capture” in the renewal offer and/or minimum increases), a common mistake is to use exactly those parameters in their budget model. This inadvertently implies that there will be absolutely no negotiation on renewals. We’ve typically found that, even in companies with policies against negotiation, it’s still unwise to assume for budget purposes that renewal offers will never be adjusted downward.

There you have it! Either build or buy budget applications with the following characteristics:

• Budget at the unit type level, not the community level
• Use drivers as inputs with occupancy being an output
• Model seasonality for occupancy drivers and for rent drivers
• And make appropriate reductions in your renewal capture and minimum renewal increase parameters to reflect the reality that some percentage of renewal offers will be adjusted