No sector has reacted more severely to the recent concern about interest rates than the Mortgage REITs. Higher long-term rates are a two-edged sword for leveraged Mortgage REITs like Annaly (NYSE:NLY) and American Capital Agency (NASDAQ:AGNC). Higher long-term rates reduce the value of their mortgage portfolio and thus the book value of the shares. The other side of the two-edged sword is that higher long-term rates and lower prices of mortgage securities provides an opportunity for Mortgage REITs to reinvest the monthly principal payments they receive in higher yielding mortgage securities. This fact has been underappreciated by the market. However, a highly leveraged Mortgage REIT with say 9 to 1 leverage and CPR of 11% would be generating new cash available for reinvestment from prepayments of principle each year approximately equal to the entire equity of the REIT.
While rising long-term rates may be a two-edged sword for leveraged Mortgage REITs, rising short-term rates are a dagger to the heart. The real risk to a highly leveraged Mortgage REIT is that shot-term rates will rise. Higher short-term rates generally mean smaller spreads between what a leveraged Mortgage REIT earns from its portfolio and the interest it pays to finance the securities bought with borrowed funds. When short-term rates get high enough the yield curve can actually become inverted. That is why most of the hedging done by leveraged Mortgage REITs involves swaps, swaptions and Eurodollar futures positions which attempt to mitigate the effects of a possible increase in short-term interest rates.
Rising short-term rates are even worse for leverage-on-leverage ETNs like MORL that borrow money to buy a portfolio of Mortgage REITs. Recent market activity in MORL and the Mortgage REITs would lead one to believe that an increase in short-term rates was imminent. I think such an increase is not coming any time soon.
The relevant rate for leveraged Mortgage REITs and is the repo rate that Mortgage REITs pay. Repos or repurchase agreements are essentially loans against securities with negligible credit risk such as agency mortgage securities where the legal title changes as the borrower sells the securities to the lender with an agreement to repurchase the securities at a specified date plus interest. That makes them virtually risk-free. Thus, while holders of Lehman Brothers commercial paper took huge losses. Counterparties with Lehman Brothers in repo transactions had no problems, since an inability of Lehman to buy back the securities allowed the lender to sell the collateral even during the bankruptcy proceedings.
MORL and the Mortgage REITs now pay about 40 basis points on their repo borrowing. The banks that provide repo financing are flush with cash and have over $1 trillion in excess reserves on deposit at the Federal Reserve. Repos compete with other virtually risk free instruments. If other short-term rates rise so will repo rates and visa-versa. If banks could not receive 25 basis points on reserves, they would be forced to buy t-bills and repos thus driving the rates on those instruments down sharply.
Most investors now believe three things about the Federal Reserve, money and interest rates. They think that the Federal Reserve is artificially depressing rates below what would be a "normal" level. They believe that in the process of doing so the Federal Reserve has enormously increased the supply of money and they believe that the USA is on a fiat money system.
All three of those beliefs are incorrect. One benchmark rate that he Federal Reserve has absolute control of is the rate paid on reserves deposited at the Federal Reserve. That rate is now 25 basis points, after being zero since the inception of the Federal Reserve in 1913 until recently. If the Federal Reserve had left that rate at zero t-bill rates would now be even lower than they are now. The shortest t-bills rates would now probably negative.
Paying interest on reserves combined with the subsidy to the banks of providing free unlimited deposit insurance on non-interest bearing demand deposits is keeping t-bill rates positive. Absent those policies the rate on t-bills would be actually negative. The Chinese and others all over the world are willing to pay anything for the safety of depositing funds in the USA. Already, Bank of New York Mellon Corp. has imposed a 0.13% charge on large deposits.
An investor who believes that interest rates are headed up may respond that the rate paid on reserves is a special case and that the vast increase in the money supply resulting from the quantitative easing must result in higher rates when the Federal Reserve reverses its course. The problem with that view is that the true effective money supply is still far below its 2007 level.
Money is what can be used to buy things. Historically money has first been specie (gold and silver coins), then fiat money which is paper currency and checking accounts (M1) and more recently credit money. The credit money supply is what in aggregate can be bought on credit. Two hundred years ago your ability to take your friends out to dinner depended on whether or not you had enough coins (specie) in your pocket. One hundred years ago it depended on the quantity of currency in your pocket and possibly the balance in your checking account if the restaurant would take checks.
Today it is mostly your credit card that allows you to spend. We no longer have a fiat money system. Today we have a credit money system. Just because there is still some fiat money does not negate the fact that we are on a credit money system. When we were on a basically fiat money system there was still a small amount of specie in circulation. Even today a five cent piece contains about 5 cents worth of metal, but no one would claim we are still on a specie money system.
Fiat money is easy to measure; M1 was $1.376 trillion in 2007 and was $2.535 trillion in May 2013. The effective money supply is the sum of fiat money and credit money. Credit money cannot be precisely measured. However, When the person in California whose occupation was strawberry picker and who had made $14,000 in his best year was able to get a mortgage of $740,000 with no money down and private equity could buy a company like Clear Channel in a $20 billion leveraged buyout, also with essentially no money down, the credit money supply was clearly much higher than today. A reasonable ballpark estimate of the credit money supply is that it was $70 trillion in 2007 compared to $50 trillion today.
The effective money supply is the sum of the traditional fiat money aggregates plus the credit money supply. Thus, despite the clams of Ron Paul and Rick Perry to the contrary, the effective or true money supply has fallen drastically over the last few years.
The decline in the total effective money supply is why the recovery from recession has been so sluggish. Those who think interest rates have only one way to go may be in for a surprise. This is especially true for short-term rates that are the keys to the prospects for the leveraged Mortgage REITs.
Now that the Federal Reserve has indicated that it will be reducing its purchases of longer-term treasuries and mortgage securities, if the economy were to stumble, a face-saving way for the Federal Reserve to react would be to reduce the rate paid on reserves deposited at the Federal Reserve. That would remove an incentive for banks not to make loans and lower all risk-free rates. The repo rate paid by leveraged Mortgage REITs would most likely fall by 25 basis points which would increase the spread and yield on an 8-to1 leveraged REIT by 200 basis points.
Right now it appears that the market is focused only on the risks that higher interest rates pose to the mREITS. Will Rogers famously quipped that investors should be more concerned with the return of their money than the return on their money. That may usually be good advice. However, an increase in long-term rates while short-term rate remain unchanged would result in higher spreads and yields for the leveraged mREITS. The last time that long-term rates were at today's levels was about a year ago, short-term rates were the same as today and the prices of leveraged mREITS like NLY and AGNC were much higher than today.
With the leveraged mREITS investors should consider the return on their money as well as the risks to the return of their money. If one looks at the extreme case of MORL, which is now yielding about 33%, it would take less than three years of short-term rates remaining at current levels for the cumulative dividends on a compounded reinvestment basis to exceed the purchase price. Thus, after three years an investor would be "playing with the house money" and would have a positive total return even if the shares were to go to zero after the third year.
Furthermore, not only has the Federal Reserve made it clear that there will be no increase in short-term rates until some period after the quantitative easing ends, but there is a chance that the Federal Reserve could lower the rate it pays on reserves. That would set off a scramble for competing riskless instruments such as t-bills and repos which would significantly reduce the leveraged mREITS borrowing costs.
Of course if the unemployment rate does not fall to 7.0%, even the tapering of quantitative easing could be postponed, and if it does not fall to 6.5% any tightening of short-term rates would be put off indefinitely. To the extent that the increase in long-term rates we have just seen impacts the real economy, especially housing, that must be considered a likely scenario. That would make the purchase of leveraged mREITS today a compelling bargain.