PrefBlog

Sheila Bair of the FDIC has testified to the Financial Crisis Inquiry Commission:

In the 20 years following FIRREA and FDICIA, the shadow banking system grew much more quickly than the traditional banking system, and at the onset of the crisis, it’s been estimated that half of all financial services were conducted in institutions that were not subject to prudential regulation and supervision. Products and practices that originated within the shadow banking system have proven particularly troublesome in this crisis.
…
As a result of their too-big-to-fail status, these firms were funded by the markets at rates that did not reflect the risks these firms were taking.

I don’t think it’s as cut-and-dried as that, unless she’s talking about the GSEs – which she might be. If investors underestimated risk – or estimated it correctly but got caught on the wrong side of a bet – it does not follow that they did so in the expectation of a bail-out.

This growth in risk manifested itself in many ways. Overall, financial institutions were only too eager to originate mortgage loans and securitize them using complex structured debt securities. Investors purchased these securities without a proper risk evaluation, as they outsourced their due diligence obligation to the credit rating agencies.
…
Consumers and businesses had vast access to easy credit, and most investors came to rely exclusively on assessments by a Nationally Recognized Statistical Rating Agency (credit rating agency) as their due diligence. There became little reason for sound underwriting, as the growth of private-label securitizations created an abundance of AAA-rated securities out of poor quality collateral and allowed poorly underwritten loans to be originated and sold into structured debt vehicles. The sale of these loans into securitizations and other off-balance-sheet entities resulted in little or no capital being held to absorb losses from these loans. However, when the markets became troubled, many of the financial institutions that structured these deals were forced to bring these complex securities back onto their books without sufficient capital to absorb the losses. As only the largest financial firms were positioned to engage in these activities, a large amount of the associated risk was concentrated in these few firms.

Much of this is just fashionable slogan-chanting, but it is interesting to see how the rating agency problem is cast: in terms of investors outsourcing their due diligence rather than as evil rating agencies inflating their output. This is an encouraging sign, putting the onus squarely on the investors – in stark contrast to Angelides ill-advised remarks yesterday, which implied that securities firms have a responsibility to sell only products that go up.

The GSEs became highly successful in creating a market for investors to purchase securities backed by the loans originated by banks and thrifts. The market for these mortgage-backed securities (MBSs) grew rapidly as did the GSEs themselves, fueling growth in the supporting financial infrastructure. The success of the GSE market created its own issues. Over the 1990s, the GSEs increased in size as they aggressively purchased and retained the MBSs that they issued. Many argue that the shift of mortgage holdings from banks and thrifts to the GSE-retained portfolios was a consequence of capital arbitrage. GSE capital requirements for holding residential mortgage risk were lower than the regulatory capital requirements that applied to banks and thrifts.

This growth in the infrastructure fed market liquidity and also facilitated the growth of a liquid private-label MBS market, which began claiming market share from the GSEs in the early 2000’s. The private-label MBS (PLMBS) market fed growth in mortgages backed by jumbo, hybrid adjustable-rate, subprime, pay-option and Alt-A mortgages.
…
These mortgage instruments, originated primarily outside of insured depository institutions, fed the housing and credit bubble and triggered the subsequent crisis. In addition, the GSEs – Fannie Mae, Freddie Mac, and the Federal Home Loan banks, were major purchasers of PLMBS.

In conjunction with her deprecation of investor acuity, this is a very interesting observation indeed!

During the 1990s, much of the underlying collateral for private-label MBSs was comprised of prime jumbo mortgages—high quality mortgages with balances in excess of the GSE loan limits. During this period, the securitizing institution would often have to retain the risky tranches of the structure because there was no active investor market for these securities.
…
However, the lack of demand for the high-risk tranches limited the growth of private-label MBSs. In response, the financial industry developed two other investment structures—collateralized debt obligations (CDOs) and structured investment vehicles (SIVs). These structures were critical in creating investor demand for the high-risk tranches of the private-label MBSs and for creating the credit-market excesses that fueled the housing boom.

With these high ratings, MBS, CDO, and SIV securities were readily purchased by institutional investors because they paid higher yields compared to similarly rated securities. In some cases, securities issued by CDOs were included in the collateral pools of new CDOs leading to instruments called CDOs-squared. The end result was that a chain of private-label MBS, CDO and SIV securitizations allowed the origination of large pools of low-quality individual mortgages that, in turn, allowed over-leveraged consumers and investors to purchase over-valued housing. This chain turned toxic loans into highly rated debt securities that were purchased by institutional investors. Ultimately, investors took on exposure to losses in the underlying mortgages that was many times larger than the underlying loan balances. For regulated institutions, the regulatory capital requirements for holding these rated instruments were far lower than for directly holding these toxic loans.

The crisis revealed two fatal problems for CDOs and SIVs. First, the assumptions that generated the presumed diversification benefits in these structures proved to be incorrect. As long as housing prices continued to post healthy gains, the flaws in the risk models used to structure and rate these instruments were not apparent to investors. Second, the use of short-term asset-backed commercial paper funding by SIVs proved to be highly unstable. When it became apparent that subprime mortgage losses would emerge, investors stopped rolling-over SIVs commercial paper. Many SIVs were suddenly unable to meet their short-term funding needs. In turn, the institutions that had sponsored SIVs were forced to support them to avoid catastrophic losses. A fire sale of these assets could have cascaded and caused mark-to-market losses on CDOs and other mortgage-related securities.

OK, so we’ve identified two problems:

• regulated institutions can reduce risk-weightings by repackaging, and
• regulated institutions are “forced to support” their off-balance sheet sponsored products

Unfortunately, her proposed solution actually exacerbates the problem:

For instance, loan originators and firms that securitize these loans should have to retain some measure of recourse to ensure sound underwriting.

Interestingly:

Looking back, it is clear that the regulatory community did not appreciate the magnitude and scope of the potential risks that were building in the financial system.

For instance, private-label MBSs were originated through mortgage companies and brokers as well as portions of the banking industry. The MBSs were subject to minimum securities disclosure rules that are not designed to evaluate loan underwriting quality. Moreover, those rules did not allow sufficient time or require sufficient information for investors and creditors to perform their own due diligence either initially or during the term of the securitization.
…
Many of the structured finance activities that generated the largest losses were complex and opaque transactions, and they were only undertaken by a relatively few large institutions. Access to detailed information on these activities—the structuring of the transactions, the investors who purchased the securities and other details—was not widely available on a timely basis even within the banking regulatory community.

Repeal Regulation FD!

In the mid-1990s, bank regulators working with the Basel Committee on Banking Supervision (Basel Committee) introduced a new set of capital requirements for trading activities. The new requirements were generally much lower than the requirements for traditional lending under the theory that banks’ trading-book exposures were liquid, marked-to-market, mostly hedged, and could be liquidated at close to their market values within a short interval—for example 10 days.

The market risk rule presented a ripe opportunity for capital arbitrage, as institutions began to hold growing amounts of assets in trading accounts that were not marked-to-market but “marked-to-model.” These assets benefitted from the low capital requirements of the market risk rule, even though they were in some cases so highly complex, opaque and illiquid that they could not be sold quickly without loss. Indeed, in late 2007 and through 2008, large write-downs of assets held in trading accounts weakened the capital positions of some large commercial and investment banks and fueled market fears.

In other words, regulators failed to ensure that the trading book was, in fact, trading and failed to apply a capital surcharge on aged positions.

In 2001, regulators reduced capital requirements for highly rated securities. Specifically, capital requirements for securities rated AA or AAA (or equivalent) by a credit rating agency were reduced by 80 percent for securities backed by most types of collateral and by 60 percent for privately issued securities backed by residential mortgages. For these highly rated securities, capital requirements were $1.60 per $100 of exposure, compared to $8 for most loan types and $4 for most residential mortgages.

Like the market risk rule, this rule change also created important economic incentives that altered financial institution behavior by rewarding the creation of highly rated securities from assets that previously would have been held on balance sheet. For example, as discussed earlier, the production of large volumes of AAA-rated securities backed by subprime and Alt-A mortgages was almost certainly encouraged by the ability of financial institutions holding these securities to receive preferential low capital requirements solely by virtue of their assigned ratings from the credit rating agencies.

In other words, it wasn’t just investors who were outsourcing their due diligence – they were joined by the regulators.

The federal housing GSEs operated with considerably lower capital requirements than those that applied to banks. Low capital requirements encouraged an ongoing migration of residential mortgage credit to these entities and spurred a growing reliance on the originate-to-distribute business models that proved so fragile during the crisis. Not only did the GSEs originate MBSs, they purchased private-label securities for their own portfolio, which helped support the growth in the Alt-A and subprime markets. In 2002, private-label MBSs only represented about 10 percent of their portfolio. This amount grew dramatically and peaked at just over 32 percent in 2005.

Good! A return to the role of the GSEs!

A reserve fund, built from industry assessments, would also provide economic incentives to reduce the size and complexity that makes closing these firms so difficult. One way to address large interconnected institutions is to make it expensive to be one. Industry assessments could be risk-based. Firms engaging in higher risk activities, such as proprietary trading, complex structured finance, and other high-risk activities would pay more.
…
The largest firms that impose the most potential for systemic risk should also be subject to greater oversight, higher capital and liquidity requirements, and other prudential safeguards. Off-balance-sheet assets and conduits, which turned out to be not-so-remote from their parent organizations in the crisis, should be counted and capitalized on the balance sheet.

I like this part, it’s good stuff!

It’s a pity she didn’t develop her attack on the GSEs further: it seems apparent that they were the kings of the too-big-to-fail castle and had very low capital requirements. But, perhaps, she simply wants to lay the groundwork for somebody else to bell the cat.