How increasing replacement costs imply a widening investment moat

Experienced real estate investors know to keep an eye on replacement cost when considering rehab deals.

The idea is to try to ensure your property will have a cost advantage vs. its neighbors.

The thought process is pretty simple: When considering doing a project, you want to look at what it would cost a competitor to buy a lot nearby and build a building from scratch to compete with you. You want to try to ensure that your all-in cost, on a per square foot basis, will be lower than your competitor.

Here’s an example of the calculation:

  1. Assume you’re looking at a 10,000 sq ft, 10 unit building you want to buy and rehab
  2. Say you’ll be all in for $3,500,000, which equates to $350 / sq ft
  3. Assume further that the lot next door is zoned for, say, five units, and similar lots have sold for $1,000,000 (eg $200,000 / unit of land)
  4. Assume that building a 5,000 sq ft, 5 unit building ground up on that lot would cost $300 / sq ft, or $1,500,000

To come in and compete with you, someone would need to spend $2,500,000 ($1,000,000 for the lot, then $1,500,000 to build). So, for his 5,000 sq ft building, he would be all-in for $2,500,000 / 5,000 = $500 / sq ft.

To carry the calculation to its conclusion:

  1. Assume you can each get $3,000 / month for your units ($36,000 / unit / year) and that expenses will equal $10,000 / unit / year, implying net operating income per unit of $36,000-$10,000 = $26,000
  2. On your 10 unit building, you have invested $3,500,000 to get $26,000 x 10 = $260,000 of net operating income, or a yield of $260,000 / $3,500,000 = 7.4% (awesome!)
  3. Your competitor would be looking at investing $2,500,000 to get $26,000 x 5 = $130,000, or a yield of 5.2% (a terrible outcome)


Obviously, the higher the cost to compete with you goes, the lower the yield a competitor can expect. And the lower the yield he can expect, the less likely he is to come in and compete with you.  Less supply means more pricing power for existing suppliers, implying higher rents and, therefore, net operating income and free cashflow.

What we have here is a naturally-occurring investment moat (for more on the concept, you go read Warren Buffett), one which works to protect and enhance your return over time.

Want to end by sharing a few implications of the above:

  1. If the city wants the market to supply more apartments, it desperately needs to reduce the cost of doing so. Most of the conversation in this area lately has been about up-zoning, which in theory ought to reduce the per unit cost of land. Much more attention needs to be paid to construction costs, which, after all, make up a much large portion of the total cost of building ground up.
  2. As construction costs have gone through the roof over the past few years, the moat protecting existing owners has widened.

The numbers are out of whack

Our business model allows us to generate yields which are consistently 200 basis points in excess of “market”.

In other words, if any random ding dong can buy a 4% cap, then we can reliably create a 6% by doing what we do.

But, right now, a 6% isn’t that great.


Well, interest rates for multifamily loans are ~4.5% / year.

So, if you borrow $2MM, you’re looking at annual debt service (inclusive of principal pay-down) of ~$121,150.

$121,150 / $2,000,000 = 6.1%.

That means, the more you borrow against your 6% yield, the worse the cash-on-cash yield gets. Ouch.

We really need to be making 6.5% yield deals to make the math work… but prices right now make that hard to achieve.

Eventually, if interest rates keep rising, prices are going to come down, which will allow us to make higher yield deals and bring the math back into alignment.

But you can wait a long time for an irrational market to come to its senses!

Why the Fed’s growth forecast matters

If you follow macro-economic news, today was a big day.

In addition to raising short-term interest rates, the Federal Reserve bumped up its estimate for economic growth in 2018, from 2.1% to 2.5%.

The question for LA landlords is whether this increased growth rate will result in additional jobs / wage growth.

Why should landlords care?

Employment growth is the single most important factor driving rents. It’s pretty obvious why… more people with jobs means more people looking for apartments. More people looking for apartments means more demand, and more demand without additional supply means higher prices (eg rents).

Wage growth is probably the second most important factor. Over the past several years, rent growth in core LA has outpaced wage growth by a pretty large margin… like 5-8% vs. 2%. To the extent that wage growth accelerates, landlords ought to be able to keep pushing up rents, too.

We have a bunch of smaller projects going into lease-up in January / February. Will be interesting to see what effect the “animal spirits” loose in the economy have on our lease ups.

The math behind what we do

At this point in the cycle, when we consider a new deal, we spend a lot of time thinking about leverage.

Mainly, we’re looking at how our pro forma unlevered yield (eg the cap rate we’re trying to hit post renovation) compares to the projected interest rate on the refinance we’ll do at that point.

I know this is a little boring, but stay with me!

Let’s look at three scenarios, all $5MM all-in deals.

Scenario 1: 7% unlevered yield

  • Invest $5MM
  • Hit 7% unlevered, so $350k in Net Operating Income
  • Building appraised at a 4.5% cap, implying value of $7.8MM (divide $350k by 4.5%)
  • Borrow 65% LTV at 4.5%
  • That’s a $5.1MM loan, with annual debt service of $309k
  • So, we’ve got all the money invested back out of the deal and we’re earning $350k-309k = $41k / year
  • Levered yield of $41k / $0 = Infinity!

Scenario 2: 6.25% unlevered yield

  • Invest $5MM
  • Hit 6.25% unlevered, so $313k in Net Operating Income
  • Building appraised at a 4.5% cap, implying value of $7.0MM (divide $313k by 4.5%)
  • Try to borrow 65% LTV at 4.5%, but can’t, because the NOI won’t exceed the debt service by enough to make the banks (or us!) comfortable
  • Instead, borrow $4.3MM (61% LTV), with debt service of $260k
  • Leave $5MM-$4.3MM = $700k in the deal
  • Receive free cashflow 0f $313k-260k = $53k
  • Levered yield of $53k / $700k = 7.6%

Scenario 3: 5.75% unlevered yield

  • Invest $5MM
  • Limp into a 5.75% unlevered, so $288k in Net Operating Income
  • Building appraised at a 4.5% cap, implying value of $6.4MM
  • Borrow $3.95MM (ouch), with debt service of $239k
  • Leave $5MM-$3.95MM = $1.05MM in the deal
  • Receive free cashflow 0f $288k-239k = $49k
  • Levered yield of $49k / $1.05MM = 4.7%

As you can see, in Scenarios 1 & 2, the more you borrow, the better the deal gets. That’s because the unlevered yield exceeds the cost of the debt (which is the total debt service divided by the loan amount). In Scenario 3, because the cost of the debt exceeds the unlevered yield, the more you borrow, the worse the deal gets.

The above is, in a nutshell, why our business exists. We get paid to deliver unlevered yields that are in excess of the cost of the debt, allowing investors to benefit from leverage, rather than get killed by it.

Fixing the small lot ordinance

Was listening to a podcaster bemoan the failure of the Small Lot Subdivision Ordinance to deliver affordable housing in LA.

There’s a lot of noise around this issue, because the homes that have been built under the ordinance have ended up being pretty expensive.

Have two, distinct points to make about this issue:

  1. It is insane to compare brand-new small lot homes to older conventional homes in the same neighborhood and complain that the small-lot homes aren’t priced at a discount. Of course they’re not… they’re brand new! (By the way, over time, I expect the prices of small lot homes to fall in real terms relative to similarly-sized conventional because they don’t come with much land.)
  2. The reason it’s impossible to deliver small lots at “affordable” prices is the parking requirement.

The idea behind small lot is to cram a bunch of free-standing homes onto one lot.

The problem is that the parking code mandates two parking spaces per unit. That’s a ton of square-footage – usually 18′ x 9′ for one car and 15 x 9′ for the other. That’s ~300 sq ft of ground-floor space. Then, you need ~10′ of backup for each space… another ~330 sq ft (the rest of the backup is generally in the driveway outside the home). That’s a total of ~630 sq ft of your ground floor building envelop devoted to parking two cars.

Given the way the ordinance works, the total footprint for these homes is going to be roughly 35′ x 30′ or something… so, 1050 sq ft on the ground floor.

But 630 of that space is devoted to a garage, leaving you just 420 sq ft, which needs to support an entry hallway and stairs up, leaving just enough room, if you’re lucky for a bedroom or office space on the ground floor.

And that means having to fit two bedrooms, two baths, a kitchen and living space into the 1050 (less stairwell) on the second floor.

How do developers respond? By going up another story, to fit more habitable space onto the same footprint. And the result of the third story is considerably more square footage and higher construction cost, leading to “luxury” pricing.

Want to fix the small lot ordinance? Reduce parking required to one space, then limit the number of stories to two. The result would be a bunch of more efficient, cheaper homes.

Which was kind of the point, right?