Description

I am recommending the common stock of New York Mortgage Trust (NYMT) for investment based on:  

1) Its substantial discount to tangible book value (approximately 58% appreciation is possible from NYMT’s closing market price on 8/14/20 of $2.79 to the most recently reported book value of $4.40); 

2) the likelihood that its book value will continue to appreciate over the coming quarters; 

3) the likelihood that the stock will ultimately appreciate to at least 90% of its future book value; 

4) its 20 cent annualized dividend (7.2% at the current stock price of $2.79); and,

 5) the likelihood that this dividend will increase as the company’s income grows over the coming quarters.

NYMT is a mortgage focused real estate investment trust (mREIT).  NYMT entered 2020 as a “hybrid” mREIT, and, through the carnage visited upon all mREITs during the COVID-related disruptions beginning in March, has transitioned to a purely credit-oriented, non-agency mReit.  For members unfamiliar with this jargon, I’ll briefly summarize what the terms agency, non-agency, and hybrid refer to in this context:

Agency mREITs invest primarily in the securities of Government Sponsored Enterprises (GSEs), such as Fannie Mae, Freddie Mac, and Ginnie Mae.  Because the market views the GSEs as having an implicit guarantee from the US Government (which was tested and effectively proven correct during the housing crisis of 2008-2009), they are regarded as having extremely limited credit risk, and are accordingly priced at relatively tight spreads to US Treasuries under most circumstances.  As such, agency mReits will usually leverage their portfolios between six and ten times to achieve the low double-digit returns on equity that their investors typically seek.  The tight spreads also mean that agency mREITs tend to finance their portfolios with cheaper, though riskier, short-term, marked-to-market funding.  The combination of high leverage and price sensitive funding requires careful management of the agency portfolio’s duration, because agency mortgages carry a risk known as negative convexity, which means that the value of the portfolio can decline both when interest rates rise (like typical bullet maturity bonds) and (unlike bullet bonds) when they fall.  This is because mortgages can be prepaid at par and most agency portfolios are comprised of securities that are on average priced above par.  

Non-Agency mREITs invest in securities and whole loans backed by mortgages on single family homes, multi-family structures, commercial real estate, and other developments that do not have the implicit government guarantees enjoyed by agency securities.  Because these products involve accepting varying levels of credit risk, they offer higher yields per unit of duration and require less leverage to achieve the return objectives of their investors.  Since many of these securities and loans are priced at or below par, negative convexity is less of concern for managing these portfolios than overall credit risk.  While many non-agency mREITs couldn’t resist using short-term, marked-to-market financing fairly extensively prior to the disruptions in March, most have been transitioning to more expensive financing sources that aren’t market sensitive after learning the lessons of this recent near-death experience. 

Hybrid mREITs invest in both agencies and non-agencies.  Management teams typically allocate capital between the two strategies based on the relative returns available between each.  Going into March, there were no mREITs I followed that characterized themselves as exclusively non-agency.  Those that chose to have credit exposure also held agencies in their portfolios.  After the March maelstrom hit, some hybrids sold their credit portfolios and became solely agency mREITs, and some sold their agencies to become solely non-agency mREITs. 

During this past March, both the agency and the non-agency mortgage markets effectively went bidless, requiring all mREITs to raise capital against what they had financed using marked-to-market facilities at the same time.  A number of mREITs asked their financiers for forbearance through the crisis, but I am not aware of any that escaped substantial, permanent book value impairment from being forced to sell securities that, while mostly all still known to be “money good”, were priced at shockingly low prices due to the temporary supply/demand imbalance.  While the Federal Reserve stepped in very quickly with a cut in the federal funds rate to zero and a massive incursion to purchase agency securities, the liquidation was too rapid to even save the solely agency mREITs.  

Below is a summary of the book value destruction for agency and hybrid mREITs between 12/31/19 and 3/31/20:

MARCH BOOK VALUE DESTRUCTION:

As might be expected, the hybrids fared far worse than the agency mREITs, as agencies enjoy much better liquidity and the Fed’s efforts were directly supportive of agencies.  NYMT fared better than the average for the hybrids, though that number is skewed somewhat by the very large hits endured by Invesco (IVR) and AG Mortgage (MITT), which both had relatively high leverage and a larger percentage of their non-agencies financed with price sensitive repo facilities. 

As mentioned above, through this crisis, a number of the hybrids apparently made the decision to move either towards the credit direction by selling their liquid agencies as those prices quickly began to recover, or towards the agency direction by selling their credit portfolios, presumably with the expectation that while prices were clearly poor at the time, they were likely to get even worse.  Most of the hybrids, including Chimera, Invesco, MFA, AG Mortgage, and New York Mortgage, sold their agencies and went the credit route.  Two Harbors sold their credit book and held on to most of its agencies.

From today’s standpoint, I prefer to look at the mREITs with credit exposure for investment.  This is because 1) the underlying securities for the credit portfolios still have substantial room to recover, and thus the potential for book value improvement is much higher; 2) the discounts to most recently reported book is larger for most of these non-agency names; 3) the income and dividends are suppressed by extraordinary expenses that should last just one or two quarters, such as forbearance issues and arranging new financing; 4) the credit businesses run with far less leverage than the agency business; and, 5) credit risk strikes me as far more manageable over cycles and crises than duration risk on a highly levered portfolio of negatively convex securities.  I believe that what happened in March is unlikely to happen again for the foreseeable future because virtually all of the credit managers are moving away from marked-to-market financing.

The recovery from the March quarter provides important insight into how each of these mREITs managed through the crisis.

JUNE BOOK VALUE RECOVERY

Here again we see the agencies did better on average.  Not only did they hold up better in the crisis, but they appear to have recovered more quickly coming out.  Some of this is to be expected, as agency prices still had some room to improve in April and May.  The credit securities really only began improving in late May and early June.  Two Harbors crystallized its credit losses in March and holds a sizable servicing portfolio, which continued deteriorating in price as mortgage rates continued to decline during the June quarter (servicing rights move down in value as rates decline and refinancing activity increases).  

Invesco and Chimera seem to be outliers in performance, and I wonder whether Chimera was able to accurately reflect the full price deterioration in its portfolio in its March report.  Looking at the June quarterly performance, I believe the agency mREITs have little room left for price improvement from a spread perspective, and I would avoid the credit names that experienced further deterioration during that period.

Next we look at current market prices as compared to the last reported book value:

UPSIDE FROM CURRENT MARKET PRICE TO LAST REPORTED BOOK VALUE

Among the agencies, we see only single digit upside to book value, with the exception of Armour Residential.  I’m not sure why that is such an outlier, but for the others I don’t think there is enough upside from market to book, nor from where the underlying securities are priced to full recovery to justify investment in a highly levered, negatively convex portfolio of low yielding agency securities.

Within the non-agencies, Anworth, MFA Financial and NY Mortgage seem investable, whereas AG Mortgage remains a bizarre outlier at a substantial premium to book despite management’s exceptional value destruction during the crisis.  I’m a bit wary of Chimera as well, given the June quarterly performance.

I chose NYMT for my recommendation because of the management team’s long, successful history and the way they navigated through the crisis.  Coming into 2020, NYMT had generated a 313% total return over the prior 10 years, outperforming the Bloomberg Real Estate Investment Trust Mortgage Index by 106 percentage points.  In the depths of the crisis, NYMT used a bounce in the value of its agency portfolio as a source of liquidity so it could hold on to its much more deeply depressed credit portfolio (the opposite of Two Harbor’s approach).  NYMT met all of its margin calls without having to resort to outside expensive, dilutive financing while its stock was trading at a distressed price (which Chimera and MFA chose to do) to keep its credit portfolio. 

I could go further into a number of esoterics regarding NYMT’s portfolio, including its diversification across single family and multi-family, its securities vs. whole loan exposure, weighted average duration, average FICO score, etc., but I will leave that for q&a, if members are interested.  All of this is available on their latest earnings presentation, and while I am satisfied with the portfolio’s diversification and its focus on residential rather than commercial exposure, I can’t really add much other than to say that the management team’s track record gives me confidence that they own these securities at attractive prices that are likely to appreciate as we move through the economic recovery. 

In its most recent earnings presentation, NYMT disclosed that it has an additional 95 cents per share of recovery value embedded in its portfolio.  Adding this to its recently reported $4.40 book value gives a potential upside to $5.35 per share.  Over the last 5 years, NYMT has traded at an average price/book of .985.  If it trades to .9 times this potential future book value, we will see a 91.7% return from the current price, plus a 7% dividend yield that is highly likely to increase as we wait.

Catalyst

Continued recovery in non-agency security prices, increase in income, dividend increase, increase in book value, collapse in market price to book value discount.