Wednesday, February 13, 2008

The Mortgage Mess: Who is to Blame?

The subprime mortgage problems are discussed, virtually daily, in the nightly news, and remain front page news in the printed and internet media. Almost universally, the press traces the problems to the greed of mortgage lenders and Wall Street investment bankers. A recent Sixty Minutes episode on CBS, for example, intimated that lax lender underwriting standards had drawn unwitting would-be homeowners into purchases and mortgage structures they could not afford. Worse yet, the default rates were highest among minority and central city borrowers.

Charges of predatory lending--intentionally targeting those who could ill-afford financial adversity--now fly fast and furiously. Cities with concentrations of low-income and protected class borrowers (such as Cleveland and Baltimore), now threaten lawsuits against lenders and others. Fingers point at bankers, appraisers, the Wall Street investment bankers and rating agencies.

But is this really the whole, or even the most important part, of the story? Are we mesmerized by the fire and forgetting to look for the matches?

CRA: Policy Match Number One?

The irony in the predatory lending story is that, not much more than a decade ago, the popular, political view of mortgage lenders was very different. In the 1980s and 1990s, the widely-held view was that mortgage lenders discriminated against the poor and stifled urban redevelopment by refusing to lend in run-down neighborhoods. It was widely alleged that mortgage lenders discriminated against women and minorities, and against central city neighborhoods. (The latter form of alleged discrimination was called "redlining".) Indeed, the Community Reinvestment Act, passed in 1977, memorialized these suspicions in law:

The CRA was enacted in 1977 to prevent redlining and to encourage banks and thrifts to help meet the credit needs of all segments of their communities, including low- and moderate-income neighborhoods.

The implementing regulations of the CRA gave bank and thrift regulators broad power to penalize banks that did not appear to be lending to "underserved" mortgage borrowing segments. In so doing, it pressed regulated lenders into doing things they might not have otherwise done.

CRA Pressures Mount

Mortgage lenders came under much greater CRA scrutiny in the 1990s. President Clinton was so convinced of lender bias that he even dispatched "testers" to pose as vulnerable home buyers to catch the bankers in the act of discrimination. His Comptroller of the Currency, Eugene Ludwig, sent fleets of inspectors into banks, looking for disproportionate patterns of lending. Threatening the banks with fines and sanctions, Ludwig bragged in 1997:

HMDA data from 1993 to 1996 show increases in mortgage originations for Hispanic Americans and blacks of 56 percent and 55 percent respectively, more than three times the 14 percent increase for white borrowers. Similarly, the rate of increase for low income borrowers was more than one and half times the rate of increase for middle and upper income borrowers. And, from 1993 to 1996, home loans in low- and moderate-income geographies increased 33 percent, while gaining only 21 percent in upper-income geographies. Our anti-redlining efforts have clearly begun to pay off.

Note that it is exactly the borrowers whose gains are ascribed to CRA who are, today, suffering financial distress. The Office of the Comptroller of the Currency, even today, summarizes the effect of the CRA as follows:

Under its impetus, banks and thrifts have opened new branches, provided expanded services, adopted more flexible credit underwriting standards, and made substantial commitments to state and local governments or community development organizations to increase lending to underserved segments of local economies and populations. (Emphasis added.)

A recent retrospective study of CRA's effect, done by Federal Reserve System economists, confirms that CRA regulation had an effect: "...a majority of surveyed institutions engaged in some lending activities that they would not have in the absence of the CRA.." However, the same study found that a "significant minority of institutions incurred losses conducting some of their marginal CRA-related lending activities." In other words, lending to marginal borrowers under pressure of CRA was not financially viable.

The Industry Evolves

Many economists believe that the mortgage discrimination assumed by the CRA was never a demonstrated reality. If one accepts this view, the CRA and its implementing regulations thus constituted a selective tax on the profits of regulated mortgage lenders. The special sanctions and additional costs imposed on regulated mortgage lenders undoubtedly provided some impetus for lending to move into unregulated capital markets. Several of my client banks, in fact, sold their mortgage subsidiaries.

At the same time the pressures on the regulated entities were mounting, the non-bank capital markets developed improved mechanisms for marketing mortgage loans directly to investors. With the costs of operating a bank stripped away, origination and interest rate costs fell. The strategy was to bundle loans into tradable securities (a process called securitization).

The CRA increased pressure on the asset-backed securities market. Indeed, CRA was modified in 1995 specifically to facilitate securitization of CRA loans. The very first CRA loan-backed securitization was offered in October 1997. The $384.6 million offering was touted as "...promoting home ownership in traditionally underserved markets through a comprehensive line of competitive and flexible affordable mortgage products." Ironically, the giant Bear Stearns investment bank was party to this transaction. Bear Stearns would come to rue its involvement in subprime lending when its stock plummeted 93% on rumors of subprime losses and it was sold to JP Morgan in March 2008 in a deal orchestrated by the Fed.

Securitization does, indeed, offer the opportunity to deepen lending to poorer credits. With securitization, mortgages did not need the high origination costs of brick-and-mortar banks and thrifts, nor did the features of the mortgages need to match those required by conforming loan, secondary market players such as Fannie Mae and Freddie Mac. In addition, the underlying cash flows of the individual mortgages could be stripped into "tranches" or different levels of risk. This allows tailoring to the needs of investors, broadening investor interest and lowering the cost of funds overall.

The secruitization phenomenon, in turn, fed back on the types of underlying mortgages that could be issued. Riskier types of mortgages could be issued, because the investor pool was no longer limited by the relatively staid instrumentation of conforming mortgages. The non-prime mortgage instrumentation, in turn, made mortgage credit available, albeit at a price, to low-quality credits--buyers with impaired credit or employment histories, etc.

Coincident Trends in Housing and Fed Policy

At the same time that CRA pressure and capital market innovation were deepening the penetration of mortgage credit, the underlying asset that secured the mortgages (housing) was rising rapidly in value. The two trends were undoubtedly related. That is, greater credit availability increased the quantity of funds chasing scarce housing (especially site) supply.

In addition, however, President Clinton raised the effective rate of return on housing assets with his promotion of legislation in 1997. This legislation allowed an individual to take $250,000 in capital gains out of the sale of a home, free of capital gains tax, every other year. This, combined with the effectively high leverage afforded by mortgage finance, made housing an extremely attractive investment class. It is not coincidental, in my view, that the origins of the home price "bubble" can be traced back to the 1995-1997 time period.

The Fed played a crucial role, too. As the 1990s came to a close, fears of Y2K disturbances prompted the Federal Reserve Board to flood the economy with liquidity. This suppressed interested rates, and fueled both the housing and dot com asset price appreciation. When Y2K proved a non-event, the Fed abruptly withdrew liquidity, and the frothy stock market (and may dot com dreams) crashed.

What happened next, in my view, was prologue to the current mortgage market blues. Having see-sawed the credit markets once before and after Y2K, the Fed abruptly lowered the Fed Funds rate by nearly 400 basis points, to only one percent. The Fed held that rate low for nearly four years, creating a a steeply sloped yield curve. Opportunities to "borrow short and lend long" suddenly were very profitable.

What better place to implement a borrow-short-and-lend-long strategy than in the mortgage market? Especially since the household portfolio reallocation after the dot com equity decline was further increasing demand for housing assets.

The Fed Lights the Second Match?

The steep yield curve lasted nearly four years (a policy period referred to as the "Greenspan Put"). Low cost credit, and rising home prices, made buyers, lenders, investment bankers, rating agencies, and investors alike, complacent.

The securitization of a pool of mortgages provides investors in those securities with credit risk diversification, but not protection from systematic risk. That is, if the odd borrower defaulted, or a housing market got weak her or there, mortgage-backed securities would buffer the bad loans with the good, the weak markets with the strong. But if something affected all markets in the same direction, the values of the underlying mortgages, their housing asset security and securities backed by the mortgages would move synchronously--down.

The shock that did it was the abrupt end of the Greenspan Put. The Fed Funds rate was raise as quickly and as far as it had been depressed after 2000. By 2005-2006, the yield curve was flat or negatively sloped, and the short-long opportunity was gone. In addition, the advantage of a security backed by a diversified pool of weak credits became a liability. Diversity become synchrony, and the pooling and stripping of the mortgage credits became a source of opaqueness.

Wth no way to easily reverse the stripping and pooling of mortgage debt, counterparty risk became nearly impossible to appraise. In these conditions, market participants assume the worst, creating further illiquidity and bearishness.

Who is to Blame?

Bull and bear markets exist because market momentum is a powerful contributor to investor expectations. Fundamentally, all market participants deserve some blame for running so consistently with the herd. In econometrics, we say that expectations were formed in an overly-simplistic, auto-regressive way--i.e., that the past would always predict the future.

But there were two policies that, in my view, constitute the "matches" in this scenario. One is the intervention of the CRA regulations. These regulations were tantamount to a mandate to make low-quality loans. The second match was the see-saw monetary policies of the Federal Reserve. The tendency of the Fed, under Greenspan, to manage soft landings, and to engineer real-sector recoveries, is a mistaken policy. Most soft landings become hard landings, and volatility in financial markets is amplified in an attempt to moderate real cycles.

Ironically, Greenspan understood that the "democratization" of mortgage credit carried risks; in a speech in San Francisco as early as 1997, he expressed concern about subprime mortgage trends. Unfortunately, like so many in the regulatory environment, the political correctness of CRA and lending to weak credits, coupled with his focus on micro-managing the economy, apparently made him turn a blind eye.