The first months of 2023 have dramatically increased volatility in the financial sector. The collapse of multiple US regional banks and the near-collapse of Credit Suisse (CS) has created jitters across the leveraged financial sector. Financial companies have never experienced such a significant and rapid increase in interest rates as they had last year. This quick change has led to a reversal in the US M2 money supply, meaning liquidity is leaving the financial system. As liquidity levels decline, leveraged banks, equity investors, and non-bank lenders (mortgage REITs, etc.) must sell assets simultaneously, opening the door to cascading losses as asset sellers vastly outnumber available buyers.
Fundamentally, the Federal Reserve can do little to stop the reversal, as significant stimulus would likely exacerbate existing inflation issues. As long as real interest rates remain as high as they are while the yield curve is so inverted, the liquidity environment will likely become further constrained. See below:
These data present a relatively clear macroeconomic backdrop facing financial institutions. High real interest rates (or rates after inflation) lower the fair value of most assets, including bonds, commercial real estate, and most equities. Inverted yield curves cause short-term borrowing rates to rise disproportionately to long-term investment returns, causing negative pressure on net interest margins. Finally, the Federal Reserve’s ongoing net sales of Treasury bonds caused some of the “cash” created in 2020 to be removed from the financial system, straining deposit levels and forcing many banks to sell assets. Since real rates have risen quickly, most financial institutions are racing to sell holdings at losses. In my view, those losses will likely accelerate as virtually all financial institutions (banks and non-banks) are looking to sell while relatively few have the leverage capacity to buy.
The commercial real estate market is likely the most significant under-discussed risk factor facing the broader financial market. The US commercial property market is enormous, and many banks and non-banks have made substantial loans to the sector. Many of these loans were created around 2020-2021, when ultra-low real interest rates spurred a buying spree in the property sector, propelling commercial property prices to extreme highs. Now that real interest rates are much higher, many commercial properties are being sold at growing losses. These losses have created “knock-on” issues as investors race for the exit and spur liquidity issues in certain REITs. As one example, Blackstone’s (BX) huge $70B “BREIT” saw redemptions increase that the company recently blocked withdrawal requests, meaning it may need to sell properties to shore up liquidity quickly.
One company that has exceptionally high exposure to this issue is Arbor Realty Trust (NYSE:ABR). ABR is extremely popular among many income investors due to its double-digit yield and strong performance from 2018-2021. The company benefited tremendously from the sharp decline in real interest rates (and, therefore, cap rates) during that period. However, its prospects have crumbled over the past year as it faces falling property prices. I wrote a bearish article on the company last summer, and it has declined by ~27% since. Despite this, I remain one of the only analysts with a negative outlook on the company, as many seem to overlook its apparent risks due to the attractiveness of its high dividend. Looking forward, I agree that ABR is less overvalued today and could see a return to stability; however, given the magnitude of the strain in the CRE sector, I still believe ABR is likely headed lower in the long run.
A Sharp Decline in Commercial Property Prices
The total value of US commercial real estate was estimated to be around $21T in mid-2021 when capital rates were reaching a minimum. During the 2021 sales boom, capitalization rates plummeted to an all-time low of 5.4%. Considering the historical relationship between capitalization rates and real Treasury bond yields (or inflation-indexed yields), average capitalization rates should rise to around 7.4% to make commercial properties reasonable investments compared to inflation-indexed Treasury bonds.
The full effects of this change have not been realized as the commercial real estate market has a 1-3 year lag to the financial market fundamentals. However, the shift from record-low real rates to relatively high real rates indicates that average commercial properties will likely lose around 27% of their 2021 peak valuations (based on est. cap rate changes), potentially causing a ~$5.6T total devaluation of all US commercial properties. Of course, the total value of commercial properties was estimated to be closer to that level (~$15T) in 2018. Since there has been no net increase in total property square footage since then, it is not unreasonable to assume property valuations should return to the 2018 level as the market no longer benefits from 2020-2021’s meager real interest rates.
Arbor is heavily focused on multifamily bridge loans. Problematically, because multifamily investments are “lower risk,” they usually carry the lowest capitalization rates in the commercial property sector, often closer to 4% at 2021-2022 minimums. However, lower rates cause more significant duration risks on prices, as a 2% increase in cap rates to 6% would cause prices to decline by an expected ~33% (given flat NOIs). Indeed, multifamily properties are losing value faster than most today. Accordingly, many of Arbor’s borrowers may face greater asset value exposure to changes in real interest rates. This relationship greatly benefited the firm when real rates were low in 2019-2021, but the opposite is true today.
Arbor’s Significant Exposure to CRE Prices
Arbor Realty is unique among mortgage REITs because it focuses on short-term bridge loans to commercial multifamily properties. At the end of last quarter, 98% of its loans were a bridge, and 91% were in the multifamily sector with decent geographical diversification (some focus in the Southeast). The weighted-average months to maturity was 20.6, and the weighted-average loan-to-value was 76%. That LTV level is on the high end of the typical range for CRE loans, indicating a higher risk of loan losses in the event of a rapid decline in commercial property prices. The company was also operating at high leverage, with total liabilities-to-assets of 82%.
Arbor’s strategy works well in most circumstances. At the end of 2022, its assets carried a total weighted-average yield of ~6%, while its weighted-average financing cost was ~4% (10K pg. 43). Its ~2% net interest margin is compounded dramatically due to its ~4.4X leverage ratio. The sharp rise in interest rates has minimal negative impacts on the company since its assets and liabilities have similar term lengths. Further, since the company also operates a servicing portfolio, it has a small benefit from rising rates since MSR’s rise in value with lower refinancing expectations from higher rates. That said, the benefit is offset by its originations business which has slowed due to high rates.
The company’s strategy gives it low exposure to long-term trends in the commercial property market but immense exposure to short-term declines in property prices. Bridge loans are riskier because borrowers are usually not qualified for CMBS loans. Most usually seek to refinance at much lower CMBS rates at the end of the bridge loan maturity after pursuing property upgrades. Some borrowers seek bridge loans to temporarily raise liquidity from properties to fulfill other projects or obligations. Thus, should property prices decline over those two years, LTV levels will likely rise to levels outside the typical CMBS maximum of 75%. Considering Arbor’s loans have a weighted-average LTV of 76% and property values are declining, it seems likely that many of its borrowers will not be able to repay obligations over the coming two years.
What is ABR Worth Today?
In my view, we’ve only begun to see the impact of higher real interest rates on the commercial property market. The commercial property market has a significant “lag effect,” so investors must take a forward-looking view instead of focusing on backward-looking measures (TTM EPS, book value, etc.). If I were only looking at Arbor’s 2021-2022 performance, I would agree with the consensus view that the stock is terribly undervalued and is a significant dividend investment. However, since I look for forward-looking indicators (such as real interest rates) and consider the cascading risk factors inherent to the banking sector, it seems Arbor is at high risk of suffering considerable asset impairment or liquidity strains.
The core issue in commercial property is relatively simple. CMBS lenders require LTV’s generally be below 75%, so if property prices decline, many of Arbor’s borrowers will be unable to secure new financing to repay their bridge loans. Commercial property prices are finally beginning to drop to reflect higher real interest rates and could fall by as much as 40% this year under a recessionary scenario. Defaults on commercial real estate loans have already reached a 14-year high despite a relatively small decline in property prices. Transaction volume has collapsed, just as many borrowers and lenders face more significant liquidity needs, implying 2023 may see sellers vastly outnumber buyers, causing a rapid decline in property prices.
The impact these changes may have on Arbor’s net asset value is challenging to measure. That said, if its loans have a W/A LTV of 76%, and the importance of those properties declines by an expected 20% (though more may be possible), then those LTVs would rise to ~95%, making the majority of Arbor’s borrowers unable to secure CMBS or similar financing. Arbor currently has an allowance for credit losses of below 1%, which I believe is very low considering today’s economic and monetary reality. It will take significant time for the property-price impacts to be realized in loan losses; however, given Arbor’s leverage, I believe the company is at material risk of losing most of its remaining book value over the next two years as its borrowers are unable to repay loans in-full due to property price declines.
Arbor is trading at a discount-to-book of 12% today. The stock declined recently due to a short report questioning its management decisions and potential misuse of capital and auditing issues. While investors may want to take note of that report, I generally do not focus on such speculations. I am primarily bearish on Arbor due to economic and monetary issues that are, at this point, outside of the company’s control. However, one point that I believe investors should consider that is written off by bullish analysts is the fact that many, if not most, of the company’s loans may not be able to be turned into agency loans at maturity due (Agency takeout model) due to falling prices (and sharply rising LTVs).
The Bottom Line
It is challenging to model the impact that may have on Arbor’s equity. Still, I do not believe a 50% decline is off-the-table due to its leverage and the extent of the potential deterioration in multifamily prices over the next two years. In other words, I would not personally buy ABR until its price is half of its current book value, or around $6.50. However, I would not short the stock since these impacts may take too long to realize them. Due to the apparent likelihood of a crisis this year, the US government may take measures to shore up capital or reduce covenants in the CMBS market. Until then, Arbor should continue to generate decent cash flows, creating high carry costs for short-sellers.
Still, at the very least, I believe ABR investors should not overlook its risks due to its abnormally high yield. Free lunches are rare in financial markets, and in almost all cases, stocks with double-digit yields carry an increased risk of suffering immense equity declines, given problematic circumstances. Arbor’s “Achilles Heel” is a sharp and rapid decline in multifamily prices, and, in my view, based on the available evidence, 2023 may be the worst year for multifamily property prices on record. To me, the tremendous impact that may have on Arbor is great enough to offset the value of its yield entirely.