It’s hard to know what to expect from fractional real estate investing returns, especially with the additional fees that are in place. Given Arrived Homes’ popularity, we’ve decided to take a look at projecting the returns from some of their offerings to get a feel for what investors might be able to expect under different circumstances.

## Key Takeaways

- Unless there are very severe market conditions, it should be very uncommon for investors to lose principal at the end of a 7 year holding period.
- That having been said, market conditions do have a huge impact on results.
- The range of performance across 3 different scenarios was:
- 5 Years: -1.6% – 204.3%
- 6 Years: 14.9% – 294.9%
- 7 Years: 24% – 430.9%

- The average total return from our base case was:
- 5 Years: 20.9%
- 6 Years: 50.7%
- 7 Years: 69.5%

- Given the range of outcomes, it’s important to look into the details to better understand what you feel might be a realistic expectation for returns.

## Methodology

For the analysis I manually picked 15 different Arrived Homes offerings to review. I wanted each offering to be in a different real estate market and I also tried to gather a mix of newer and older offerings, as well as those with and without leverage.

The basis of the model is then calculating investor’s starting positions after the initial fees, extrapolating the growth in the housing value over time, extrapolating dividend payments over time, and factoring in fees and costs as well.

The values used to model appreciation change in each case and will be detailed in their individual sections. Arrived properties use interest-only mortgages, so in cases where there is leverage, there is no mortgage pay-down being modeled into the returns.

For dividends, I used either the real values to date or the projected values in the offering. From there, we modeled a 3% annual growth of the dividend applied every two years when leases should refresh. In some scenarios we assume only a portion of these are received. Based on what we saw in the 2022 tax documents, we calculate dividends as *lowering* cost basis rather than adding on top of appreciation.

Lastly, for fees, we include the sourcing fee and we also used a 7% disposition/sale fee of the property. That is included in the model at all times, even from year 0. The basic idea there being that even if Arrived decided they needed to sell the offering the day after it closed, they would still incur closing costs. Thus the true “value” of the shares is always lowered by these would-be costs.

Note that I do include cash reserves and property improvements at face-value as part of the “value” of a share. Trying to model out the changes in these values is complex and unlikely to be correct, so we simply include the base values to the bottom line in all years.

## Bad Case

Let’s start with the bad news first. In this scenario we assume that from year 0 to year 1 there is a 10% real estate correction. From there, we model a slightly above-average annual property growth rate of 5%.

This seemed somewhat fair given all markets appreciated more than that over the past 8.25 years (per Zillow’s home index). Additionally, the rate of single family home appreciation has generally been increasing due to structural tailwinds. For reference, since 1991, single family homes appreciated an average of 4.3% with the rate being 4.7% since 2020 (per Credit Karma).

For dividends, we take our base calculation and reduce them by 20% as an attempt to cover a broad-range of impacts in a simple way.

Here’s what that ends up looking like for the 15 offerings we selected:

At a high level, here’s the breakdown:

- The minimum returns were at 5Y: -1.6%, 6Y: 14.9%, 7Y: 24%
- Average total returns were at 5Y: 8.4%, 6Y: 22.8%, 7Y: 39%
- The maximum returns were at 5Y: 29.7%, 6Y: 59.6%, 7Y: 96.4%

You may also notice that there’s really two distinct clusters of properties in the projection. One cluster pulls down to about -20% after the correction while the others draw down closer to 50%. The bounce back and subsequent returns of the -20% cluster is also generally lower. This is the difference between leveraged and non-leveraged properties.

In general, even in this case no one really loses principal, but the returns are mostly underwhelming. For reference, a 7% annual compound return over the exit time periods of 5-7 years would yield about a 40-60% total return. So, an average return of 39% undershoots that considerably. It’s actually even below 5%, which would be 40.7% of total return after 7 years.

## Base Case

In this scenario, we mostly project a regression to the mean. Annualized appreciation returns to the average of 4.3% since 1991 and stays there for all 7 years. We also reduce the projected dividends by 15% to account for a wide range of factors including potential falling and consolidating rents after a number of years of growth.

Here’s what that translated to with the 15 different offerings:

Which gives us this high-level breakdown:

- The minimum returns were at 5Y: 20.9%, 6Y: 30.2%, 7Y: 40.6%
- Average total returns were at 5Y: 34.4%, 6Y: 50.7%, 7Y: 69.5%
- The maximum returns were at 5Y: 79.1%, 6Y: 116.9%, 7Y: 165.6%

As with the previous case, it’s possible to visually identify two clusters – the leveraged and non-leveraged properties. The offerings without leverage basically have higher floors and lower ceilings, not falling as much, but not growing as fast either. The average return in this case tracks fairly closely to a 7% compound annual return. It underperforms that benchmark a bit at 5 years, but outperforms it a little at 7 years.

The Pinot is an outlier amongst this data set. It’s one of the earlier offerings and has a very high dividend yield. That yield is mostly unmatched on the platform and is what drives its outperformance. Excluding that, the high-end of returns is around 100%.

## Great Case

This is actually our most complicated projection and is the most “exciting” in terms of the potential for returns.

In this scenario, we used Zillow’s Home Index to calculate the annual appreciation of single family homes in that market over the past 8.25 years. That also means the rate of appreciation will now deviate between offerings.

In this scenario, we do not “hedge” the dividend yield. We assume a full 100% payout throughout the duration of the investment.

Which produces a graph that looks quite a bit different than the ones before it:

The high level stats break down to:

- The minimum returns were at 5Y: 36.6%, 6Y: 56.2%, 7Y: 76%
- Average total returns were at 5Y: 93.5%, 6Y: 131.8%, 7Y: 178.4%
- The maximum returns were at 5Y: 204.3%, 6Y: 294.9%, 7Y: 430.9%

In this configuration, the market (which the appreciation rate is based on) and whether there is leverage or not on the offering drive the returns. The Pinot is still an outlier. This is where the leveraged offerings shine the most and it’s interesting to see just how much higher their upside is.

That having been said, you can see that market selection plays a more important part here. While the non-leveraged properties are all relatively close to each other, you can see differentiation between the leveraged properties. This could also mean that *my* offering selection between leveraged and non-leveraged properties could be exaggerating the differences. For example, The Grant is the top non-Pinot performer due in large part to Atlanta’s killer growth rate. In this model, perhaps a non-leveraged Atlanta property could outperform leveraged properties in slower growth markets.

The average returns in this case look more inline with 12-16% annualized growth rate, which would be a *fantastic* return. Though the likelihood of it is fairly low. Recent appreciation is obviously skewed upwards by the run-up in price after the pandemic. It stands to reason that prices will slow down or consolidate over the next several years. That would start to weigh the average annual appreciation rate back down.

## Closing

One last area I wanted to look at before we wrapped up was the leveraged vs non-leveraged properties. I grabbed two offerings from each group with middling performance and charted their performance over our different scenarios:

It does *generally* seem like leveraged properties provide better returns, with some more risk. A tricky part of this conclusion though is that the total returns include dividends. If newer properties were leveraged, it would be at higher interest rates and, thus, with lower dividends. That would negatively impact their total returns. So, it’s less clear if leveraged properties under the *current* financing conditions would still offer as much potential for out performance. Also, this is always an imperfect comparison. We are again not comparing properties in the same markets that are true comparable sales.

Overall this was an interesting exercise that I hope was helpful. I really don’t think the “Great” case returns are very realistic. I personally expect the “Bad” and “Base” cases are more likely, with the potential for single family home appreciation to be stronger in some of the tail years.

We’ll continue to keep an eye on Arrived, their offerings, and their performance. So, stay tuned for more coverage!