Assessing CRE Exposure Across the Financial Sector

CRE worries rising. Exposure to US commercial real estate has been a source of increasing concern in the financial sector; this has been most plainly visible in pure-play vehicles such as REITs and CMBS, but it was also among the key concerns driving Deutsche Bank’s March swoon. While REITs, insurers, and banks share exposure to CRE, the degree and nature of exposure varies significantly by (and within) industry. The purpose of this report is to offer a sector-wide situation report, and to highlight some ways to dig deeper into each industry’s specific exposures.

REITs: Equity REITs account for about 10% of the roughly $20 trillion U.S. CRE market and have seen rising cap rates across all property types, though the rise has been especially sharp for the office segment. Even for office REITS, there remain some bright spots, however. Consensus EBITDA estimates still imply growth over the next two years, and their aggregate debt maturities don’t peak until 2026-27, which affords some time for market conditions to improve.

Insurance: Insurance industry exposure to CRE is skewed heavily toward life insurers, by virtue of their longer-tailed lines related to mortality risk (compared to lines such as auto and homeowners for property & casualty insurers). Life insurers’ yields on mortgage investments have exceeded those of their bond portfolios in recent years, but yields have been declining and delinquencies rising. Much exposure often comes in the form of CMBS, about which it bears noting that they are generally comprised of non-recourse loans.  

Banks: Banks are the largest holders of commercial real estate exposure in absolute terms, holding for 60% of CRE mortgage debt (excluding multifamily). Within that 60% are many cross-currents, however. CRE is a much bigger share of total loans for small banks than for large ones, yet the Fed’s 2022 stress test, which assumed a nearly 40% decline in CRE values in the extremely adverse scenario, showed much greater losses for banks operating at the high end of the market.

Why is this cycle different from all other cycles? On the liability side of the ledger, the fact that interest rates were so low from 2008-2022 suggests that the effects of higher rates on property cash flows, debt service and valuations are only just starting to be felt. Reasonable people can reasonably disagree about how high rates will go in this cycle, and about how severely market discipline was undermined by 15 years of cheap money, and those assumptions will inform estimates of where the market may find a bottom.

On the asset side of the ledger, we note two somewhat related secular trends. The first is the massive growth in private markets, and concomitant growth in dry powder, as shown in Figure 2. The second is the ongoing maturation of alternative vehicles for the wealth market, dramatically expanding the mutual access of alternative asset managers and wealth managers. The historically unprecedented capital available from these sources may help the CRE market find a bottom, albeit only after a great deal of pain. It is also possible that banks, by virtue of being better capitalized than in prior cycles, may find themselves less constrained; the bullish argument would be that the magnitude and/or duration in decline in market liquidity should be mitigated. We shall see.

REITs

Cap rates rising for all property types. Cap rates have been rising for all property types since 2021. Unsurprisingly, office has seen one of the sharpest increases, up around 300bps. As a rule of thumb, assuming flat NOI, every 100bp increase in cap rate would reduce total real estate value by something on the order of 10% - 20%. (FactSet Cap Rate Time Series Template)

Publicly traded Equity REITs are about 10% of the total CRE market. Equity REITs collectively have approximately $625 billion debt on their balance sheets, and an aggregate enterprise value of around $2 trillion, which represents roughly 10% of the estimated $20 trillion commercial real estate market. (FactSet US REIT Full Comps)

Equity REITs’ average interest cost was 4.6%, and NOI to interest coverage ratio was 4.5x. With US CMBS yields currently exceeding 6% (per the Bloomberg US CMBS 2.0 Index), both metrics face meaningful headwinds, even before figuring in the impact of a potential recession on NOI.

Among all property types, office faces the most severe fundamental headwinds. Over the next two years, however, consensus estimates project growth in total EBITDA, even for the office sector.

While debt maturity has been an elevated concern for CRE, publicly traded Office REITs do not face significant near-term debt maturities; the peak remains relatively far off in 2026 and 2027. With current weighted average coupon rates at or below 4% across maturities, however, they could face significantly higher interest costs if they refinance. (FactSet Industry Debt Maturity and Interest Cost Analysis)

Insurance

Aggregate P&C real estate investment exposure of $4.5 billion. Over the last five years the insurance industry has maintained a steady level of investment in real estate and mortgage loans (both residential and commercial). Both the life and P&C industry’s real-estate investment as a percentage of admitted assets has remained in the single digits (Figure 7) over the last five years that ended 2021. Although this allocation is a small percentage of investments for each company, in aggregate the P&C industry’s balance sheet reflected about $4.5 billion of real estate investments in 2020 and 2021 (Figure 8).  

Life insurers’ exposure is higher than P&C. The life insurance industry invests a much higher percentage of assets in real estate than the P&C industry given an allocation to mortgage loans. The split between real estate owned and mortgage loans for the life insurance industry is shown below (Figure 9). This difference in allocation from the P&C industry reflects the longer duration of life insurance mortality liabilities compared to the P&C industry’s shorter-duration liabilities for auto and homeowner claims. The difference in the two real estate investment profiles is driven by the ALM, or asset liability matching, process.  

Insurers’ mortgage yields have been rewarding, but declining. The yield produced by mortgage investments has been slightly higher for the life insurance industry than bond investments, although yields have been falling. The yield of both bond investments and mortgage investments has fallen slightly over the last five years (Figure 10). 

Figure 11 presents a breakdown of the life insurance industry’s total mortgages, loans by type, as well as various percentages to capital and reserves. Delinquencies and foreclosures have increased the last two years but remain well below 1% (Figure 9). Commercial loans account for approximately 90% of the growing total mortgage loans.  

Company level analysis vs industry level. The Summary Investment Schedule found within the Statutory Statements provides data to analyze the allocation of the investment portfolio to mortgage backed securities. Figure 12 is a sample of group level data analyzed using FactSet’s Insurance Real Estate template US Insurance - Real Estate Holdings, which can be found in the template library. The template shows an insurance company’s allocation of the bond portfolio and total portfolio between residential and commercial mortgage-backed securities, at a group level, as well as a maturity distribution of the securities. 

Subsidiary level detail includes real estate and mortgage loans owned. In addition to the group level data above, FactSet’s Insurance Real Estate template pulls a subsidiary level list of individual properties shown in Figure 13 and mortgage loans in Figure 14 from Statutory Statements. 

Banks

A tale of two bank industries. Banks hold the majority of CRE mortgages by volume, with 61% of total CRE debt residing on their balance sheets (excluding multifamily). Within the industry, however, there is a dichotomy between large and small banks. As shown in Figure 15, CRE accounts for 24% of total industry loans outstanding, but CRE represents just 13% of total loans at large banks, while accounting for 44% at small banks. Figure 16 shows this at the individual bank level; the list of banks with the highest CRE/Total Loans ratios is dominated by small- and mid-sized banks. 

Large bank – small bank divergence widening. That divergence has been widening; a year ago, CRE loans were 14% of large banks’ total loans, and only 40% of small banks’ loan books. Figures 17 and 18 tell the tale; though year-over-year CRE loan growth was 13%, CRE loans grew just 3% in the past year at large banks, with the growth rate generally bouncing around the low single digits, while at small banks, CRE loans grew 18%, and that growth rate has been accelerating.

All CRE exposure is not created equal. While smaller banks have more aggregate exposure to CRE, the exposure at larger, wholesale banks tends to be riskier. That is a broad generalization; banks of all sizes will rise and fall with the underlying strength of their footprint, and their underwriting discipline. Plenty of small and mid-sized banks have managed to blow themselves up with foolish underwriting decisions, some more flamboyantly than others. The Federal Reserve’s stress tests provide a window not just into overall risk but that dichotomy. In the 2022 stress test, under the severely adverse scenario, CRE values decline by 40% over 2 years (the same assumption is included in the 2023 stress test as well), and the aggregate losses for the banks tested represented 9.8% of total CRE loans, but Morgan Stanley and Goldman Sachs were outliers, with projected loss rates more than double the peer group. Meanwhile, JP Morgan Chase, with a wholesale business complemented by a national retail and middle market franchise, saw projected losses of less than half the peer group.

Key leading indicator suggests the worst is yet to come. Each quarter the Federal Reserve conducts the SLOOS—Senior Loan Officer Opinion Survey—in which it polls said loan officers on whether they are tightening standards, and seeing increasing demand for, loans of various types. This data set has some value as a leading indicator, and recent results have not been encouraging. As shown in Figure 20, both indicators have reached levels only previously registered during recessions.

Historical perspective on CRE credit quality. While there is plenty of cause for concern, about both CRE as an asset class, and banks’ exposure to it, Figures 21 and 22 offer some historical perspective. Both the noncurrent loan rate and net charge-off rate for bank CRE loans remain at historically low levels. Last quarter, the aggregate net charge-off ratio for CRE increased sixfold—to 0.06%. Of course, another way to read those charts is to infer that we may be on the cusp of an extremely sharp rise.

2008 is not the measure of all crises. The graphs offer one other insight that seems worthy of note. Over the past 15 years the 2008 financial crisis has, justifiably enough, been used as a quick and easy proxy for a worst-case scenario. It is often referred to as the worst banking crisis since the Great Depression, and by many measures it was. Not so for CRE, however. Note that for CRE credit, the peaks of 2008 fell well shy of what the industry endured in 1989-91.

FactSet colleagues Gary Lu, Stewart Johnson, and Lee Li contributed to this article.

This blog post is for informational purposes only. The information contained in this blog post is not legal, tax, or investment advice. FactSet does not endorse or recommend any investments and assumes no liability for any consequence relating directly or indirectly to any action or inaction taken based on the information contained in this article.