# Failed Math: A Surety Bond Simulation

--

“Knowledge is anything that increases your ability to predict the outcome.”

~Michael Lewis, The Undoing Project

It’s no secret to multifamily that surety bonds have failed as a broad-based financial instrument to replace security deposits [1]. On its face, bonds are excellent at creating affordability and converting leases — creating “liquidity” as bond providers champion. A seductive proposition for sure. Yet the other half of the story remains — evaporated claims funds leaving properties exposed to bad debt.

Historically, claims funds have been separated and managed by property (pooled). Recently, new bond formats have come to market that do not pool by property (unpooled) but instead at the portfolio or carrier level — yet the failed math equation remains. In fact, unpooled formats introduce new business risk, as accountability is removed from the individual property with claims funds commingled across entire portfolios. It’s not a story of pooling, it’s a story of math.

I will run a simple simulation to make three important points for operators to consider:

1. The math behind surety bonds leads to eventual failure — regardless of whether the bond is pooled or unpooled [1].

2. Using a bond value pegged to the average deposit [2] and the NAA reported national average move-out loss [3], we use math to conclude that bonds depend on unfiled claims — also known as breakage — to remain solvent over time.

3. Using standard reinsurance performance indicators, we can further conclude that bond programs with insufficient breakage will sustain loss ratios exceeding 100% and risk nonrenewal by their reinsurance carriers [4].

# Failed Math

“If you can’t explain it simply, you don’t understand it well enough.”

~Einstein

Bonds are “insurance-like” products that give residents the option to bypass the required deposit and instead pay a fraction of the deposit as a non-refundable fee (premium). Properties may then draw against the premiums — the claims fund — to cover losses typically for unpaid rent and damage, up to the limit of the bond value [5]. The claims fund is only as large as the contribution from collected premiums. This is the fundamental structure of the surety bond instrument— regardless of how claims funds are organized.

What goes into the claims fund:

• Residents pay a one-time fee of around 17.50% [5] of the bond value, typically equal to the value of the required deposit.
• We find an average deposit of \$423.50 [2] based on our analysis of over 500,000 rent ledgers. We will round the bond value up to \$500.00, making the cost of the bond \$87.50. Some providers break up the one-time fee into monthly installments.
• Bond providers then deduct a service fee — typically up to 40.00% [5] — which equates to \$35.00, in this example.
• After the \$35.00 service fee is deducted from the \$87.50 bond fee, the remaining \$52.50 is deposited into a claims fund [5], which is intended to pay for future losses.

What comes out of the claims fund:

Using the provided example, we see that an average claim of \$52.50 will reduce the claims fund balance to zero over time — \$52.50 credited, \$52.50 debited.

Here’s where the math fails.

Using statistics published in the 2019 NAA Survey of Operating Income & Expenses in Rental Apartment Communities, the nationwide average collections loss is \$103.00 per unit [3]. With \$52.50 going into the claims fund, and \$103.00 coming out to cover the average move-out loss — the claims fund depletes over time eventually becoming insolvent. The math fails regardless of whether the claims fund is pooled or unpooled.

I should also note, the NAA reported \$103.00 move-out loss [3] is after deposits have been exhausted — so the actual loss is greater when deposits haven’t been collected (as would be the case with your typical bond).

Zooming in further, we compiled data of actual collections loss (bad debt) across a handful of Top 50 NMHC portfolios — analyzing a statistically-significant cohort of over 500,000 lease move-out ledgers. We found the average collections loss to be \$326.00 per unit [6] — much higher than the figure reported in the 2019 NAA survey.

# Breakage Model

~Georg Cantor

So how do bond providers exist at all?

There are a few factors at play. But the short and quick explanation is called breakage. Breakage is simply when a claim that could or should have been filed is not [7].

“Breakage is defined as any type of service that is unused by a customer that has already been paid for in full.”

When operating against national averages, bonds can become highly dependent on breakage to survive [1]. By definition an average is an average, and though there are properties that collect more premium or have less bad debt — with an average \$52.50 premium going in, and an average \$103.00 needed to come out — a model built on breakage is revealed.

Bonds that remain solvent over time are likely to have larger than average premiums, lower than average bad debt, or contain properties that are not filing claims [1].

Breakage aside, there are a few more “features” that prolong the eventual depletion of the bond claims fund.

1. Natural time delay. Funds build a positive claims balance for an initial ramp-up period as new move-ins occur. After a ramp-up period of about 24 months based on the national average turnover rate [7], move-outs occur at the same rate as move-ins and the fund balance starts a slow decay toward full depletion.

2. Collections. Collections are run against residents for claims paid out to the property [5]. These collections efforts are often very low-yield and surprise residents, compounding an already untenable situation with a negative customer experience — typically absorbed the form of damaging online reviews.

3. Bond inflation. Providers may suggest inflating the bond value higher than the required deposit to allow more cash flow into the claims fund. This results in an increased cost to the resident, offsetting the affordability the product intended in the first place. For example, a bond value set to \$2,000.00 will cost the resident \$350.00, using the 17.50% rate.

4. Increased rates. Some bonds have a “risk-based” rate card [8] — at times approaching 50.00% of the bond value — based on the applicant’s FICO score. This is a chase from behind as the effective average rate must be doubled to cover the NAA reported collections loss of \$103.00 per unit to make the claims fund break even. For example, a bond rate set to 35.00% for a bond value of \$500.00 will deposit \$105.00 into the claims fund.

These various tactics are emblematic of a product struggling against math. Bonds are destined for failure based on the national averages cited for this simulation [1]. There is not enough premium collected to keep claims funds solvent over time.

# Reinsurance Loss Ratios

No one can see a bubble. That’s what makes it a bubble.”

~The Big Short

Spoiler alert! There is no magic insurance fairy. Even the most sophisticated multifamily owners and operators have fallen victim to a false sense of security.

Let me explain.

Bond providers issue policies that are backed by multi-billion dollar rated A (Excellent) reinsurers. This means that any outstanding claim against a policy will be paid up to the full limits of coverage — regardless of the claims fund balance [9].

Feels safe, right?

It shouldn’t. Reinsurers are profit-seeking enterprises like any other business. There is no special math with insurance that magically creates profit from thin air, as many would hope.

As standard business practice, reinsurance carriers evaluate every product they back — including multifamily bond providers — on an annual basis at minimum. Reinsurers use a simple equation called “loss ratio” to evaluate performance and make renewal decisions [10]:

Loss Ratio = Claims / Premiums

If claims paid out exceed premiums collected, the loss ratio is over 100% and the reinsurance carrier is losing money. In fact, many property & casualty reinsurers target loss ratios closer to 40–60% [11].

Now, let’s plug the bond inputs from the example above into the reinsurance loss ratio equation:

Loss Ratio = \$103.00 Claims / \$52.50 Premium = 196%

That’s a 196% loss ratio. Not good.

As a profit-seeking enterprise, reinsurers have a strong motive to cancel or “non-renew” any money-losing business line [10]. Non-renewals are a standard feature built into all contracts between reinsurers and product providers. There are recent examples of new insurtech startups entering markets — such as renters and auto insurance — sustaining loss ratios over 100% and then being dropped by their reinsurance carrier.

# Summary

In summary, multifamily operators must be aware of the structural failure built into the math behind all bond formats — pooled or unpooled — that forces dependency on a breakage model to sustain solvency over time [1].

The failed math forces operators into a bad choice set: (1) be diligent in filing claims and accelerate the depletion of the claims fund, or (2) participate in a breakage model by not filing claims and have properties absorb bad debt directly.

Unpooled bonds introduce new business risk as operators cannot be sure of the behavior of other participants in the aggregate, and therefore are unaware of the performance of the claims fund as a whole. Unlike pooled formats where operators can manage the behavior at each property to keep claims funds solvent, little can be done by any single participant in the unpooled format to prevent catastrophic failure [1].

By applying national statistics to the surety bond equation and then running the simulation atop standard reinsurance principles, we can predict the ultimate failure of bond programs over time with a high degree of certainty [1].

Caveat emptor.

Derek Merrill — Founder, LeaseLock

Citations

[1] Failure or insolvency defined as a negative surety bond claims fund balance and/or loss ratios exceeding 100% using an average \$500.00 bond value, 17.50% bond rate, 40.00% service fee, and an average claim amount of \$103.00.

[2] Average deposit based on LeaseLock analysis of over 500,000 rent ledgers over prior 60 month period.

[3] National average move-out loss based on collections loss reported in 2019 NAA Survey of Operating Income & Expenses in Rental Apartment Communities.

[4] Brown RL. (1993). Introduction to Ratemaking and Loss Reserving for Property and Casualty Insurance.

[5] RealPage DepositIQ. How DepositIQ Works For Your Community.

[6] Average collection loss based on LeaseLock analysis of over 500,000 rent ledgers over prior 60 month period.

[7] Average occupancy term based on average 51% turnover rate reported in 2019 NAA Survey of Operating Income & Expenses in Rental Apartment Communities.

[9] Jetty Blog, Luke Cohler. What does it mean when a surety bond uses “pooling”?.

[10] Investopedia Loss Ratio, Adam Hayes. What Is a Loss Ratio?.

[11] Wikipedia. Loss ratio.