As a landlord or property manager, setting rents is a high-stakes exercise.  Set them too low, you might get a tenant but you are leaving money on the table.  Set them too high, you could be unnecessarily extending your vacancy loss.  And it is with this backdrop that rent setters employ various strategies and technologies to help with pricing.  Finding that equilibrium point is not easy, especially in a dynamic market like New York City. 

            Amenity-based pricing is a common way to set rents, and this article is mainly in reference to such models.  Amenity-based pricing is a methodology in which a base rent is set for a unit type of some kind, then various amenity prices/values are added in, creating the all-in rent that renters see and pay. Individual amenity prices are set by a variety of factors, including the cost of developing the amenity, the desired payback (to the landlord), and perceived value to renters. The percentage of amenity value to total rent varies a lot, but consider it to be about 5-10% of the total rent.

            Aside from landlords and property managers, renters also have a vested interest in how rents are priced, because they want to know they are getting a fair market deal, and because, naturally, it is a lot of money. 

            So we wanted to provide a few thoughts on how rents can be set and to encourage landlords and property managers to re-think the exercise in different ways. 

1.       Data vs. Emotion

Amenity-based pricing is a highly data-intensive exercise. It is all about statistically significant relationships between amenities and various renter/vacancy/lease-up data points. Put simply, if a unit with certain amenities leases more quickly than a unit without those amenities, that is a reflection of those certain amenities’ relative values.  There is a lot more to it, but that is the essence of it.

            And yet, a major known factor of the psychology of renting is that renters are not necessarily data-driven consumers – they can be emotional, irrational, and unpredictable. Especially in markets such as New York City, there is not even the opportunity to be fully data-driven and assess and digest all available factors, because events and decisions move too quickly.  A renter considers basic data such as rent, budget, number of bedrooms, and a few other data points, but they are not necessarily making or intending to make internal, completely rational statistical calculations about how certain combinations of amenities move their internal needle on renting or not. Representing their decision-making patterns within a statistical-based framework is not an accurate interpretation of what drives renters’ behaviors. At the same time, leasing managers use strategies such as reference rents and anchoring in lease negotiations, something that most renters are well aware of.  But such strategies are essentially manual overrides to the outputs of an amenity-based pricing system. So even the leasing managers on the ground recognize that amenity pricing has its limitations that must be accounted for.

Additionally, amenity-based pricing is written about a lot in the multifamily software world, and many articles are careful to point out that the system does have its limitations – an acknowledgment that is in line with what we discuss here.    

2.       Amenity Overload

Another point to reconsider in rent setting is which and how many amenities to price in. The whole point of “amenity value” is that amenities have absolute and relative values. There is an inconvenient reality in this system however that different amenities appeal to different renters. Amenity offerings need to be diverse in order to maximize value to as many different renters as possible. However, the more amenities that are added in, the more their marginal value is diminished. The system becomes overloaded with individual elements and each individual amenity loses part of its value. As a consequence, the models behind pricing get a little confusing, because what then reflects the real value to the renter? Despite the fact that not every amenity appeals to all renters, landlords and property managers are mostly pricing rents as if they do. Another way to put it: the wrong items are getting priced, and renters are paying for things they do not want, and are NOT paying for things they DO value. Such an outcome is the furthest thing from pricing optimization, which is what amenity-based pricing is supposed to do. 

Complicating this further, amenities frequently have development costs that need to be recouped. Amenity development decisions are often based on assumed return on investment schedules, which then are fed into amenity-based pricing models, which are in turn based on flawed concepts of renter behavior. The greater the uncertainty about renter behavior, the more amenities are supplied, the figurative “see what sticks” approach. Then there is more pressure to recoup the cost of an amenity that incurred a development cost, even if that costly amenity is not really wanted by renters.  These actions perpetuate the cycle and lead to diminishing marginal value.  

3.       Willingness to Pay vs. Value to Renters

As mentioned in #1 above, data scientists and rent setters use statistical relationships as a proxy for a renter’s willingness to pay and value to the renter. The logic here is that a renter’s actions, such as when exactly they decide to lease a specific unit, is a rational and reproduceable reflection of the value to them of certain features and amenities. As we describe in this article though, that’s not necessarily the case.  For a data-driven conclusion to make sense and have validity, the actor which it is describing must act rationally. Otherwise the relationships that are extracted from the data will not inform future behaviors

To determine real value, rent setters need to embrace the behavioral component of renting and look beyond physical amenities whose inclusion in amenity-based models may be motivated more by expensive development costs than having real value to renters.  Statistical significance within large amounts of data glosses over what amenity pricing is supposed to do: optimize rents at a granular level. And having the costs drive the value amounts to a post-hoc justification for incurring amenity development costs: an amenity is built, costs are incurred, then those costs are considered “value” to renters after-the-fact via methodologies that do not necessarily capture actual renter behavior and preferences. It is contrarian to see it this way, because the purpose of amenity-based pricing is to add more granularity and revenue capture into pricing decisions, but in fact it can do the exact opposite. 

Until the data-driven approach can be harmonized with a more individual behavioral component, which truly captures real marginal value to renters, pricing will continue to be subject to miscalculation.

We hope this discussion has got you thinking about how rents are set and different ways to think about improving the exercise.  We are not impugning the work of multifamily data scientists or those behind the amenity-based pricing models, as that work is extraordinarily complicated and takes extraordinary skill. We are simply commenting on the system itself and different issues pricing managers may want to consider as they strive to optimize their revenue and understand what their renters truly value.