Today's Market Observation  03.09.2010  Mon  Tue  Wed  Thu  Fri  Eckert Archive

Option ARM-ageddon

Part 1

by Richard A. Eckert, cfa | march 9, 2010

ARM-ed but under siege

By the middle of 2003, the GSEs and the capital markets (agency MBS and private ABS investors)—along with an extended period of low long-term interest rates—had rendered the savings and loan (thrift) industry nearly obsolescent. The ARMs in which they specialized—indeed, as mentioned above, the only type of mortgage they could really hold in portfolio—were increasingly eschewed in favor of the certainty and security provided by fixed-rate mortgages (FRMs). Not only did FRMs provide the peace of mind that accompanies a payment that does not vary, but, depending on the shape of the yield curve, the coupons on FRMs were not only competitive with that of a fully-margined ARM, but, at some points in time, the lower cost alternative.

The entire ARM portfolio lending model was under siege. Not only were regularly repricing ARMs uncompetitive with FRMs and hybrid ARMs—thrifts were loath to migrate to the latter because the prepayment penalties that made them work were hard to enforce in the conventional loan market, where they ran up against Fannie Mae (NYSE:FNM—$0.99) and Freddie Mac (NYSE:FRE—$1.16), which stopped buying mortgages with prepayment penalties nearly three decades ago (lenders could not impose those penalties knowing it took the two largest buyers of home loans out of the market for their mortgages)—but it was becoming increasingly difficult, if not impossible, to hold loans in portfolio. Increased volatility and the general trend downward in interest rates increased the risk of spread compression and prepayments. The latter actually reinforced the former as many ARM indices repriced with a lag and replacement loans were written at a teased rate if thrifts wanted to remain competitive at all. Additionally, any unamortized deferred origination costs had to be absorbed into net yield in the month the loan was repaid. Speaking of origination costs, the thrifts were uncompetitive there as well. The savings institutions’ brick-and-mortar infrastructure, salaried employees, and manually underwritten loans placed them at a significant disadvantage to centralized mortgage banks deploying Fannie Mae’s Desktop Underwriter® or other similar automated underwriting and pricing engines and a highly motivated cadre of commissioned account executives. 

Finally, there was the matter of capital. Paradoxically, publicly traded thrifts were put in the unenviable position of being held to prudent capital standards by their regulators while trying to generate competitive returns for their shareholders. In the independent mortgage banking space—particularly in the conforming mortgage sector, where creditors knew there was an exit strategy of last resort (and a profitable one at that) in Fannie or Freddie as long as they were confident in the lender’s ability to underwrite to the GSEs’ standards—creditors were willing to advance up to 95-99% of the acquisition cost of a loan. And independent mortgage bankers, depending in which segments of the market they lent, could turn that capital over every 15-60 days. 

There were those thrifts that argued their competition was not the independent mortgage banks—or even Fannie and Freddie—but the “end users”, the RMBS markets. Even here, the savings and loans were at a severe disadvantage. Although they were compelled to generate competitive returns for their shareholders, many RMBS investors were merely trying to duration- or horizon-match their assets and liabilities. That is, they were only attempting to match the estimated cash inflows of their mortgage-backed securities with the projected retiree income benefits, claims payments and endowment obligations they are required to meet. Because of that, they had actually become enamored of the embedded prepayment option in most mortgages as it allowed them to create customized cash flows to meet those requirements. In either case, the thrifts were between a rock and a hard place. At the origination end, they had to compete with lenders with much lower capital requirements. At the portfolio end, they had to compete with investors with much more broadly defined—and forgiving (in the sense of being truly long-term, not quarter-to-quarter)—return requirements.

To be certain, thrifts were not entirely run out of the mortgage business at this point. ARM originations did, in fact, increase 80% in 2002 and 62% in 2003¹. But the increases represented: 1) 27% and 34% hikes, respectively, in the total origination market²; 2) a rebound off of a particularly weak 2000-2001, when total origination volume doubled year-over-year, but ARM share was cut in half—to 11.25%—resulting in roughly the same volume of adjustable rate mortgages, ~$265 billion in both years³; and 3) a surge in subprime/Alt-A lending, almost all of which was hybrid adjustable (and very little of which was originated by regulated depositories). What is important to remember here is that portfolio lenders make precious little of their money in originating loans—although that was where they spent an inordinate percentage of their internal budgets (much of it buried in deferred origination costs and net yield)—but, instead, in the net interest income generated from those loans after they had been retained for investment. And, despite the huge increases in new loan production, many thrifts actually reported shrinking balance sheets in 2002-2003. They simply could not keep up with the volume of prepayments as both interest rates and refinance costs/friction declined. 

More insidiously, high levels of prepayments and the growing share of FRMs resulted in no small amount of adverse selection as the most creditworthy of both existing and new borrowers—and, in the case of existing borrowers, those with the most equity in their homes—gravitated toward fixed-rate loans. A telltale sign of the impending debacle were the boasts made to this author by the managers of thrifts he covered in the early part of this decade that they had successfully found high-quality, new borrowers just outside Fannie’s and Freddie’s lending criteria—jumbo loans, less than 20% down, with little to no income documentation, but high FICO scores (it turns out these could be easily gamed or that they captured exogenous variables like the borrower’s willingness and ability to refinance instead of the willingness and ability to repay—see Beware of Geeks Bearing Stats, 9/23/2009 Market Observation). Thus, even before the proliferation of perhaps the deadliest mortgage instrument known to humankind, the option ARM, the thrift industry was stepping briskly down the credit slope. But, as long as home values and credit scores held up, their regulators—and the elected officials peering over their shoulders to make sure they were doing all they could to boost the homeownership rate—were willing to buy off on it. 

The “perfect storm”

But 2004 represented the “perfect storm” for the portfolio mortgage lending industry. In the three years ended December 2003, U.S. home prices rose sharply, climbing 42% nationally⁴.  In some regions, they positively soared. In the Los Angeles and San Diego MSAs, home prices advanced 58% over the same three years⁵; in Miami, they increased 50% from December 2000 to December 2003⁶. At the same, interest rates began rising after hitting a cyclical low in June 2003. The yield on a 10-yr Treasury note, to which many fixed-rate mortgage coupons are benchmarked, rose almost 150 bp from its cyclical low of 3.13% in June 2003 to 4.62% at the end of June 2004, which is when the Fed began to remove its policy accommodation, pulling up the short-term rates to which ARM coupons were linked. Thus, by the end of 2004, rates were rising across the board.

Sharp increases in home values and interests combined with a flat real wage base (see Exhibit I) to create a veritable crisis in affordability, leaving many would-be homebuyers on the sidelines. The NAHB/ Wells Fargo (NYSE:WFC—$27.35) Home Opportunity Index (HOI), which had dropped below 60 only in the ’91-’92 and ’01-’02 recessions, fell to 50.5 by the third quarter of 2004⁷. In time it would decline to 40.4⁸ (see Exhibit II). Note the inverse relationship between the HOI and ARM origination volumes in Exhibit II.

Most propitiously for portfolio ARM lenders, the developing affordability crisis coincided with visible changes in borrower behavior and cultural values. Large segments of the homeowner population—and this would include speculators and income property investors as well as savvy owner-occupants—came to view their homes as investments, not the roofs over their heads. In an effort to boost their return on investment, they sought to minimize the “IC” in “ROIC”, using cash-out refi transactions to reduce their capital at risk—in many cases, to $0 or less, getting their lender and 2nd lien loan partner to finance not only 100% of the appraised value of the home, but insurance and tax escrows, origination fees, prepayment fees on the previous loan, etc.

 

Another large segment of the homeowner population fell under the sway of the siren song of the affluent lifestyle pitched to them in just about every channel of the popular media. Although the advertising of both small and large luxuries has been around just about forever, the new media, new marketing techniques, advanced data mining and predictive modeling applications, and certain shifts in cultural norms made the messages—“you can live the life of the rich and the famous” and “immediate gratification”—more sophisticated, more powerful and, yet, more subtle at the same time. A significant number of Americans wanted their “pie in the sky” and they wanted it now. Cash-out refis delivered on both criteria. They enabled the lifestyle spending that nonexistent real wage increases would not support (see Exhibit I). And baby boomers, Gen-Xers, Gen-Yers had become extremely comfortable with the debt their parents and grandparents eschewed. They had no misgivings about taking on debt and using the equity in their homes to finance lifestyle purchases. 

Thus, by late 2004, the value of thrift espoused by earlier generations had given way to a culture of profligacy. The motivation of borrowers in a refinance transaction was not so much minimizing their coupon rate of interest, monthly payment or loan term, but maximizing the number of dollars they could withdraw from the ATM their homes had become. In the case of speculators/ investors, the objective had become to minimize the number of dollars invested in their properties. And paying off the mortgage on one’s home became a quaint, dated concept. Instead of being un-indebted, the goal had become to leave not dollar one of equity un-monetized.

Lending no small amount of support to these conclusions are the data underlying Exhibits III and IV. Never had so many Americans extracted so much equity from their homes (Exhibit III).

Using a combination of cash-out refis, closed-end seconds and HELOCs, the population of prime borrowers captured in Exhibit III would pull $1.1 trillion in cash out of their homes between the end of 2003 and the end of 2007. When considered with subprime and Alt-A mortgages—the majority of which represented cash-out refi transactions—the total balances of equity monetized and withdrawn from American homes were of staggering proportions. Which helps explain the steep increase in U.S. mortgage debt outstanding this past decade—from $4.8 trillion at the end of 2000 to $10.3 trillion at the end of the 3rd quarter of 2009 (actually peaked at $10.6 trillion at the end of the 1st quarter of 2008)—and the equally steep decline in home equity as a percentage of the total estimated market value of all owner-occupied real estate (see Exhibit IV).

ARM-ed and dangerous

As the planets of low affordability and high profligacy aligned perfectly in their favor, the nation’s thrifts were ready to capitalize; they were going to seize this opportunity to become relevant again. And, as it turns out, they had already developed the perfect WMD—“Weapon of Mass (Home Value/Household Finance) Destruction”—the option ARM. Option ARMs permitted the borrower up to four payment options each month: 1) a fully-indexed, fully-margined, fully-amortizing loan payment based on a 30-year amortization schedule; 2) the same as 1) except that it was based on a 15-year amortization schedule; 3) an interest-only payment using the fully-indexed, fully-margined coupon rate of interest; and 4) a “minimum” payment using an interest rate of 1.00 – 2.25%, depending on the lender.

Because option 4) did not cover the fully-indexed, fully-margined interest payment, the difference between the latter and former was tacked on to the borrower’s UPB. The increase in loan balance is referred to as “negative amortization” and mortgages that allow for negative amortization are often called “neg am” loans. Most “neg am” loans capped such increases at 15-25% of the original UPB—i.e., when the loan amount reached 115-125% of the original loan size—at which point the payments became fully-indexed, fully-margined and fully-amortizing over the remaining term of the loan. All option ARMs—whether or not they hit the “neg am” ceiling—eventually reached this point. At the end of 5-10 years, all loans reset to a fully-indexed, fully-margined, fully-amortizing ARM over the remaining 20-25 year term of the mortgage. Unless, of course, they had been refinanced earlier.

Actually a west coast “innovation”, the option ARM had been around for a couple of decades. Prior to 2003, though, they had been deployed selectively—and almost exclusively—among the large California thrifts’ largest and most financially sophisticated borrowers (high net worth individuals purchasing/building a second or third home, income property investors, and single-family residential (SFR) real estate speculators). It was a win-win proposition at the time. The borrowers were willing to pay up for flexible payments, which helped them manage the cash flows from their businesses/ investments. And the lenders earned well above-average returns while protecting themselves with not only the meteoric increases in California SFR values, but pledges of additional collateral from many of the borrowers. In most cases, the appreciation in residential real estate prices alone eclipsed any negative amortization that may have taken place as a result of the borrowers having made minimum payments. Moreover—and perhaps best of all—the flexible payment options made option ARMs a portfolio product; the often uneven and unpredictable cash flows associated with these loans did not lend themselves readily, at least at that time, to securitization. Also, even back then, when housing prices were climbing at a double digit (%-age) per annum pace, RMBS investors tended to be nervous about negative amortization.

The “green light” from Greenspan

But even under ideal conditions and possessing the ideal product for those conditions, many, if not most, ARM portfolio lenders—the vast majority of them regulated depositories—were uneasy about rolling out option ARMs to the masses. The minimum payment option and the negative amortization feature gave rise to a tremendous potential for large increases in loan balances and payment shock. Which, in turn, gave rise to not only the potential for high rates of default and levels of loan loss, but backlash from elected officials and/or their regulators. In fact, even if the thrift industry could comfort itself about the risk of losses—these would most assuredly be contained by home prices they had convinced themselves would continue to soar through the stratosphere and into the mesosphere and thermosphere (where troubled borrowers could refinance or sell at a price that at least covered the UPB, even if it contained negative amortization)—they could not size the risk of legislative or regulatory setbacks. In fact, just a handful of these loans in the wrong hands—where the potential for misunderstanding, misuse and miscalculation were greatest—could create a huge political liability.

They needed political cover. Official sanction. Some kind of endorsement or sponsorship from an unimpeachable authority. On February 23, 2004, they got it. And from less an authority than the then Chairman of the Federal Reserve Bank, Alan Greenspan. I have excerpted below several of what I consider to be the more salient remarks from a speech given to CUNA (Credit Union National Association) by Mr. Greenspan on that date.

“Rising debt service ratios are a concern if they reflect household financial stress and presage a drop in consumption or a rise in losses by lenders…Financial institutions might be able to help some households in this regard by looking for ways that households—both renters and homeowners—can shield themselves from unexpected payment shocks.”

“One way homeowners attempt to manage their payment risk is to use fixed-rate mortgages…Homeowners pay a lot of money for the right to refinance and for the insurance against increasing mortgage payments…Indeed, recent research within the Federal Reserve suggests that many homeowners might have saved tens of thousands of dollars had they held adjustable-rate mortgages rather than fixed-rate mortgages during the past decade…”

“American consumers might benefit if lenders provided greater mortgage product alternatives to the traditional fixed-rate mortgage. To the degree that households are driven by fears of payment shocks but are willing to manage their own interest rate risks, the traditional fixed-rate mortgage may be an expensive method of financing a home.”

“Overall, the household sector seems to be in good shape, and much of the apparent increase in the household sector's debt ratios over the past decade reflects factors that do not suggest increasing household financial stress.”

The Chairman gave portfolio ARM lenders the “green light” they required. Once more, the option ARM delivered; they met many, if not all, of the specifications set forth by Mr. Greenspan. Because the borrowers controlled the payment, they could protect themselves against payment shock. Of course, this presumed informed and rational use of the option ARM product. In an ideal world, borrowers would make fully-indexed, fully-margined and fully-amortizing loan payments in a falling rate environment and interest-only or “minimum” payments in a rising rate environment. Assuming regular use of all four options, over time negative amortization would be at least offset by fully-amortizing payments. And implicit in Chairman Greenspan’s observations—especially the last few sentences of his speech, where he declared the household sector to be in “good shape”—was the premise that as long as the value of household assets (the largest of which was their homes) kept climbing, any increases in household liabilities (the largest of which was their home mortgages) arising from negative amortization or other sources would not increase “household financial stress.” 

Indeed, the belief among both the Chairman and thrift executives appeared to be that when combined with regular use of all four payment options, home price appreciation would render negative amortization a non-issue. With any measurable increases in the market value of their homes, option ARM borrowers could refinance their mortgages long before their loans hit the “neg am” cap or reset to a fully-indexed, fully-margined, fully-amortizing 20- or 25-yr ARM. In a worst case scenario, they—or their lenders—could sell their homes at a price that covered the UPB. To their way of thinking it seems that option ARMs represented the “best of all possible worlds.” Both the risk of loss and the risk of payment shock could be contained—or at least managed. And for the portfolio lenders—particularly the publicly traded ones—option ARMS yielded at least one more advantage. Regardless of which option a borrower chose, the lenders could report 1) a performing loan; and 2) a fully-indexed, fully-margined interest payment—and, to the extent that payment was not received in cash, it could be added to interest-earning balances. They could report positive net interest income and increases in assets—the source of future increases in reported net interest income.

By the time rates started rising across the board in mid 2004, every party concerned—loan makers, lawmakers, policymakers, and regulation makers—was ready to buy off on the product. Homeownership rates and home values were rising, GDP was advancing at a brisk pace, and the nation was at full employment (after adjusting for structural factors). It would have been “politically incorrect” to correct the housing bubble inflating. No one in a decision-making position wanted to be the person to “turn on the lights” at the party and ask everyone to go home.

Rich Eckert

Copyright © 2010 All rights reserved.

Please note Part 2 of 'Option ARM-ageddon' will post next Wednesday, March 17.

contact information

Richard A. Eckert, CFA | (415) 674-4996 | San Francisco, CA
Email | Observation Archive | Other FS Editorials