The Biggest Mistake I Made (Twice!) on My Best Midwest Investments
Forget the Mistake, Not the Lesson
One defining characteristic of Midwest multifamily housing stock is its age. The long decline in manufacturing employment that began in the 1970s resulted in almost no new multifamily housing construction for nearly 30 years. This means that Midwest multifamily housing stock is heavily skewed toward 1960 and 1970’s vintage construction, and the physical obsolescence of certain building components in these older assets needs to be carefully assessed. To the extent that some building components may have exceeded their useful lives, capital reserves should ideally be set aside to replace them. However, in value add investing, the ideal world does not always exist.
The first property I bought in 2006 was a great example of this. While the property had very attractive, competitive floor plans with abundant natural light, it was 1970 construction with engineered wood siding (T1-11) that had not been properly maintained, original construction aluminum-framed, single-paned windows, and balconies 30 feet above ground with 35-year old cantilevered joists in delicate condition. And I did not reserve for any of it.
Deferred Maintenance Will Not Cure on its Own if Left Alone
Even though I knew I had to fix these problems after closing, I did not set aside capital at closing to make the repairs. Why? Because I was like almost all investors who were just starting out: I was undercapitalized and I wore rose-colored glasses. In addition, the equity required to purchase this 58-unit building consumed almost all of the cash that I had, leaving very little left over for upgrades and repairs. Most important, as the largest investor in the project, I did not want to suffocate my ROE by needlessly overcapitalizing it.
I knew this was risky, given the age of the asset and the problems I knew of, but my going-in basis in the property was was incredibly low relative to replacement cost (and for good reason!). Furthermore, based on my brilliant models, I felt confident that this property would cash flow generously out of the gate, and my thought was that I could fund capital projects over time using excess cash flow.
This assumption proved to be too optimistic, and every month there were multiple projects competing for that “excess” cash flow. The window replacement project alone involved ordering 236 new windows of multiple odd dimensions. The new windows would require three months of fabrication, two semi trailer loads to deliver them, three mobile mini boxes for storage, two months for installation, and hard costs alone (excluding installation, storage & delivery) were $102,452.
I Knew I would Need to be Creative
As investment managers and property managers, we should always focus first on improving the resident experience first. Very closely behind this needs to be a careful focus on risk management. Our focus on resident experience helps the asset perform better with respect to rents and retention, and our focus on risk management helps reduce our exposure to liability from slip and fall injuries due to trip hazards and snow and ice hazards.
These priorities also help provide clarity on how to allocate scarce resources. After closing, I went to work immediately on finding additional resources for repairs and upgrades. Eventually, I discovered an Ohio energy conservation program directed toward multifamily housing and commercial real estate. My window replacement project was small in relation to the state’s investment targets, but the program was just kicking off and they invited me to apply. The application process was arduous and paper intensive, but I was eventually approved for a $102,452 loan with 3% interest-only payments for 10 years.
Once the windows were installed, the residents started making unexpected observations. The new vinyl framed, double paned windows not only made the apartments cooler in the summer and warmer in the winter, but many people loved how much quieter it was inside their apartments. This was very gratifying!
Why Do All These Incentive Programs, Including Tax Benefits, Exist?
As an aside, this project helped me understand why all these tax incentives and subsidies are directed toward real estate owners and managers. This window replacement program was relatively small, but it still involved manufacturing 236 brand new windows from scratch, two truck drivers to haul them all up from the Kentucky factory to the installation location in Ohio, 3 storage containers to store them during installation, and 5 installers who worked for two months to install them all. These projects obviously support a significant amount of economic activity, which is why they attract a considerable amount of federal (e.g., bonus depreciation) and state support. I was grateful for this, as this project was a lot to manage.
Concurrently with the window replacements, I also began replacing the worst of the deteriorating cantilevered joists from cash flow. In the process, we also perfected a replacement technique that allowed us to replace them more quickly with less disruption to residents. 10 years after closing, we had all the windows replaced, all the cantilevered balcony joists replaced, and a new rubber membrane roof installed — all without any reserves set aside up front.
Don’t Try This at Home!
We were able to do all this without any reserves primarily because we grew NOI sufficiently to execute two cash out refinancings, and we directed a significant amount of the proceeds toward remaining deferred maintenance items.
I did this again two years later with an even larger 150-unit property, although on this project I reserved $500,000 at closing for deferred maintenance. The only problem was that I was confident the $500,000 would not be enough on its own. Having been through this dance before with the windows and balconies on the 58-unit property, I knew it could be done on 150, but this project would involve an even more intricate choreography between growing rents, careful rationing of scarce reserve dollars, and hard choices on prioritizing numerous deferred maintenance items with fingers crossed while we got all the work completed. In addition, we had to contend with damage from five kitchen fires, two of which took out entire buildings.
The Lessons I Learned From All This
Ultimately, we did very well on both projects, and we still own them today. There were several things I learned, however. The biggest lesson was to be more confident and realistic with rent growth assumptions that are clearly supported by comps. I thought that using more conservative rent growth was a virtue that spoke to the viability of the overall value add business plan, but this was wrong. By using more realistic rent growth assumptions, the pro-forma ROE would have increased, and an increased ROE would have helped justify more upfront capital at closing. This would have meant getting the work done more quickly and returning investor capital sooner.
For example, even though the 150-unit project rents were almost 15% below the comps, and the market was growing at 3% per year, I assumed 2% rent growth in year 1, 6% in years 2 and 3, and then 2% thereafter. In year 5, our pro-forma gross income in year 5 was $1,030,000, and our actual year 5 gross income was just about on target. However, in year 10, after we were finally able to get all the work done, our actual gross income was $1,494,000, vs. original pro forma gross income of $1,184,000. It would have been much better for everyone to get this work done earlier in the hold period so we could be farther ahead in year 5.
Another trade off was the investor distribution profile. At year 10, we refinanced with a new 65% LTV loan and it returned 100% of investor capital, satisfied the 8% Preferred Return in full and produced a 12.25% simple per annum return — and we didn’t even sell the asset. The investment was obviously successful, but we didn’t sell because there is still more meat left on the bone. Investors are happy to own a cash flowing asset with known future appreciation potential, but it would have been an earlier home run if we could have completed all the value add projects more quickly and returned capital sooner.
In addition, distributions during the hold period were uneven and infrequent. This made investors nervous, even though we were meeting and eventually exceeding our pro-forma gross income targets. I know that at the peak of the last cycle, a number of sponsors were funding distributions with upfront capital, and some of my investors thought we compared unfavorably to those other sponsors who were effectively subsidizing distributions with investor capital. I explained what was happening, and these investors later learned that this is a very bad practice, but it made for some uncomfortable discussions early on.
Underwriting rent growth is a balancing act, and using assumptions that are too aggressive will get you into trouble very quickly. I would rather err on the side of conservatism, but being too cautious can also handcuff you to your business plan much longer than is necessary, and it will cause investor returns to be less dramatic because they are drawn out over time. I like to say that the only thing I know for certain about our underwriting is that I know it will be wrong, but I’ve also learned that being too conservative in order to account for these unknowns also has undesired consequences.