The US real estate market has experienced tremendous volatility over the past couple of years due to the ongoing economic shift. Historically low mortgage rates have sent home prices skyrocketing with a surge in buying activity among young Americans. At the same time, total US household debt has increased significantly, as have mortgage-backed bank assets. Today, the ratio of the US house price index to the average hourly wage is back above peak ~ 2006 levels, which preceded the bursting of the last housing bubble. Additionally, the average hourly CPI-measured rent-to-income ratio in the United States has declined at a relatively steady pace since the start of 2020. See below:
Note that the units in the charts above are arbitrary – what matters is that US homes have never been more expensive relative to income levels. Also, rent has been very high relative to income levels and is now falling (relative to income). Most likely, the surge in home purchases is partly due to an overall decline in demand for rental housing. Thus, I suspect house prices should stop rising and may even fall as the average rent-to-income ratio returns to historically normal levels (which makes renting a potentially better deal than buying a house).
Of course, the main overall factor driving this change has been the extreme decline in mortgage rates since 2020. In fact, relating the house price index to the average hourly income ratio and multiplying it by the average over 30-year mortgage rates (data here), we can see that the measure has been stable since around 2010. This means that the amount of income paid in interest on homes has been nearly constant since the bursting of the real estate bubble in the mid-2000s (and was much higher before). At first glance, this fact may give the impression that the United States is not in a housing bubble, since owners are not necessarily “stretching” to buy houses (apart from a handful of states).
Rising mortgage rates may shock lenders
Homes are very expensive relative to income levels, so what happens when mortgage rates rise? Given trends in construction and demographics, it seems quite likely that house prices will decline (at least on an inflation-adjusted basis) over the next few years. With many homeowners “locked in at low rates”, a drop in house prices would strain mortgage banks as “loan-to-value” levels rise.
As higher mortgage rates begin to expose these underlying issues, I suspect that major mortgage companies such as Annaly Capital (NLY) could face a dramatic downturn. I’ve been one of the few bears on Annaly since September, as detailed in “Annaly Capital: Superior Hedging Strategy, But Not All Risks Can Be Avoided”. Indeed, NLY is already down around 17% since the “mortgage spread” started to widen last June. See below:
The difference between 30-year mortgage rates and 30-year Treasury rates has now returned to historically normal levels of around 1.5%. As the Fed narrows, this is the last of its purchases of mortgage-backed securities and long-term Treasuries over the next month, this spread, along with the underlying Treasury rate, could continue to increase. Indeed, while the mortgage gap is now normalized, the 1.15% gap between the average 30-year mortgage (3.55%) and the expected long-term inflation rate (break-even rate over 10 years at 2.4%) is still well below the typical range of ~2-3%. To bring this gap back to sustainable levels, I suspect we could see mortgage rates rise another 1-1.5% as the Fed normalizes monetary policy this year.
Although Annaly has some hedge against most “typical” monetary events, a big enough shock could quickly damage the stability of the business. Based on the conclusion of research done on my previous research on Annaly:
Annaly’s value will decline in the event of a credit event, mortgage rate shock, or increased prepayments, but it would take an immense shock for Anna’s value to be permanently altered.
Annaly is a huge company that trades near perfect correlation to the broader Mortgage ETF (REM). Thus, while some geographies and segments of the US housing market are more or less exposed to rising mortgage rates, Annaly’s main risk factor is a rapid deterioration in the stability of US housing credit. In my opinion, this is more than likely given the strong potential for rising mortgage rates to depress home sales. In addition to the increased potential for a long-term glut (due to construction and demographic trends), Annaly and other mortgage REITs could struggle to recover from such a shock.
Why house prices may slide this year
Importantly, the average 30-year mortgage rate has climbed about 0.50% over the past month and is now almost 1% above its all-time low of 2.6% set precisely a year ago. The overall rise in mortgage rates has already led to a slight decline in total existing home sales:
I suspect we will continue to see existing home sales decline as mortgage rates continue to return to historically normal levels – possibly a return to the 4-5% range by year end. In particular, just over a third of home sales are new constructions (a record high). Additionally, multi-family housing starts are at very high levels, with the 12-month moving average being the highest since the 1980s. Taken together, these trends imply that total housing inventory levels in the states are rising even though US population growth is at an all-time low (0.1% in 2021) and could likely turn negative this year.
An aging population with fewer children will lead to lower demand for single-family homes for years to come. Combined with the surge in new construction, it seems almost inevitable that there will be a housing glut across the country, especially as the “baby boomer” generation looks to downsize and find less young families to sell houses to. Were it not for the hugely stimulating impact of rock-bottom mortgage rates, I suspect this problem would be more apparent and have a negative effect on the housing market. So, as mortgage rates rise at a rapid pace, it looks like we may be seeing a rapid decline in housing demand (especially relative to supply).
Buckle up for volatility
All in all, there are a few big, basic economic and financial trends that could be problematic for Annaly this year (and most other mREIT agencies). First, the rapid rise in mortgage rates given the combined increase in treasury bill rates and the spread between mortgages and treasury bills. Fortunately, about 80% of the company’s portfolio is hedged against a rise in rates and the yield curve (from the Q3 presentation p. 4). Annaly is also technically hedged against mortgage rate spreads (unlike others) through her large portfolio of mortgage servicing rights. These “MSRs” generally increase in value when mortgage rates rise, as higher rates cause a dramatic drop in demand for refinancing.
From the end of September (Q3) to today, the spread between the average 30-year mortgage rate and the 30-year Treasury bill has increased by around 50 basis points. Annaly’s latest sensitivity analysis (Q3 10-Q Pg. 78) suggests that a 25 basis point increase in the “MBS spread” (probably approximately the same as the measure I am using) would cause its value to decline net asset of approximately 10%. While they don’t provide an estimate of the larger 50 basis point spread shock that occurred, this would equate to an estimated 20% decrease in NAV if we assume linearity.
Large, quick shocks, like the recent one in the mortgage market, can lead to even greater losses in net asset value. There has also been a slight increase in Treasury rates since September 30, to which Annaly is slightly exposed. Still, I think it’s fair to assume that NLY’s book value today is probably about 20% lower than it was at the end of the third quarter, which equates to $6.70 per share (against $8.39). Notably, the decline in mortgage-backed securities began in early January and there was little change during the fourth quarter. Therefore, this decline in book value is unlikely to show up in its upcoming fourth quarter (instead of first quarter 2022) earnings report. Additionally, Annaly’s sensitivity estimates may be wrong or may have changed by the end of 2021; thus, I estimate that the net asset value per share of NLY today is in a wide range around $6.50 to $7.50.
Based on this analysis, it looks like Annaly is trading above its net asset value today by around 10-20% following the recent spike in mortgage rates and the sharp drop in the value of asset-backed securities. to mortgage claims – see (MBB). As we have only seen the beginning of the reduction in Fed MBS assets, agency MBS assets may continue to deteriorate as there is an apparent lack of private demand (understandable given their extremely low yields). That said, the mortgage spread is at roughly normal levels, so ideally volatility will slow down a bit. Still, as mortgage rates rise, I think there’s a good chance we’ll see lower home prices and sales, especially given the long-term housing glut. This could significantly increase credit risk for Annaly’s MSR and non-agency portfolio and potentially create larger losses.
Overall, I would avoid NLY today as it looks overvalued relative to its estimated net asset value. Additionally, there remains a high degree of downside potential to its net asset value, given the potentially significant downturn in the housing market due to the recent huge spike in mortgage rates. While rising rates and spreads may make Annaly’s new purchases more profitable, its 11% forward yield isn’t much of an advantage compared to its relatively large and immediate downside risks. Of course, if the Fed becomes overly concerned about the impact of the surge in mortgage lending and starts buying mortgage-backed securities again, Annaly’s value could return higher. In my view, this is highly unlikely given the Fed’s goal of slowing inflation.