Would Glass-Steagal Have Prevented the Mortgage Crisis?

by James Hanley

An Essay in Three Parts. Egregiously long, and about financial stuff, so read at your own risk.

Somehow, some way, the Head Start thread became a thread about the mortgage crisis, and beginning about here, Blaise P. begins arguing that the repeal of Glass-Steagall played a big role in the crisis. Along the way he claims that “You just don’t understand this well enough. If you did, you’d be making different noises,” as well as (to another commenter) “You don’t know what you’re talking about and I’m not going to put up with this uninformed crap.” Great theater, perhaps, but not conducive to reasoned discussion. As it happens, I think I’m not completely ignorant on the topic, so here goes.

I: Origins of the Mortgage Crisis
II: How Mortgage-Backed Securities Work
III: Glass-Steagall

I: Origins of the Mortgage Crisis
In my considered opinion, the mortgage crisis was the consequence of the interaction of several government policies. I find most people want a single-policy argument, whether it’s conservatives blaming the Community Reinvestment Act or liberals blaming the repeal of Glass-Steagall. People seem not to like thinking about the interactions of policies, but policy is much like chemistry, the mixture of multiple policies, like the mixture of multiple chemicals, can have undesired effects.

First is the federal government’s increased pressure on banks to make mortgages available to low income borrowers, beginning in the mid 1990s. This was a renewed emphasis on the Community Reinvestment Act, or CRA. Banks that did not provide sufficient numbers of loans in poorer neighborhoods risked being charged with redlining. The problem was that one of the reasons (when it wasn’t simple racism) for banks to not make loans to poorer people is because poor people often just don’t qualify–they’re not good credit risks, as explained by the Boston branch of the Federal Reserve.

Unintentional discrimination may be observed when a lender’s underwriting policies contain arbitrary or outdated criteria that effectively disqualify many urban or lower-income minority applicants.

Arbitrary criteria, sure, that’s inappropriate. But what does “outdated” mean? The Fed report explicitly encourages lenders to de-emphasize high obligation ratios (debt to income); allow creative means of meeting downpayments, including loans; not worry too much about poor credit history; deemphasize employment history; and count a broader set of income sources, even those that are temporary or potentially unreliable, such as unemployment benefits and overtime work. The emphasis on CRA worked, as home loans to poorer neighborhoods increased substantially, but it meant that banks had to get creative with their loan models to enable sub-prime borrowers to qualify.

But CRA itself is not a major cause pf the crisis. A report published by the San Francisco branch of the Federal Reserve points out that while 20% of all mortgages issued in 2005 were subprimes, “more than half of subprime loans were made by independent mortgage companies, and another 30 percent were made by affiliates of banks or thrifts, which also are not subject to routine examination or supervision.” After analyzing a sample of subprime loans issued by both CRA regulated lenders and independent mortgage companies (IMCs), the authors conclude that “loans originated by lenders regulated under the CRA in general were significantly less likely to be in foreclosure than those originated under IMCs.”

So we shouldn’t over-emphasize the CRA. If CRAs were a little under 20% of subprimes, and were less likely to be in foreclosure (but based on the report, at least half or more as likely), then CRA covered loans were somewhere between, roughly, 10-15% of the problem.

But the emphasis on new loan methods to meet the needs of subprime borrowers did create a new precedent, making subprime loans more normal, more of a standard business practice than–as it more commonly was in the past–a practice to be avoided.

By itself that would not have caused the mortgage crisis. There’s just not enough effect on the mortgage market. But this mid to late 1990s push for more subprime lending was followed by the Federal Reserve’s easy money policy in the early 2000s, as it tried to stimulate the economy out of recession. At the beginning of 2001 the Federal Funds rate was 6.5%. By 2003 it was at 1%. Mortgage interest rates fell from about 7% for a 30 year fixed-rate mortgage to about 5.9%. That doesn’t seem like so much, but more importantly 1 year adjustable rate mortgages (ARMs) fell from 6.7% to 3.75%.

The third policy was the Clinton-era decrease in the capital gains tax on homes that were owner-occupied for two years. Any time government policies change the rate of return on an investment they change investors’ strategies, and by reducing the capital gains tax on homes, but not for other investments, some amount of investment was shifted to home ownership, increasingly so as the bubble grew, making the returns from short-term home ownership even more lucrative. In fact a person could get a 3 year ARM, live in the house for 2 years, then sell it at the increased price and never have to worry about refinancing. Nobel Prize winning economist Vernon Smith was particularly harsh in his criticism of this policy, crediting the capital gains reduction with

fueling the mother of all housing bubbles…enabling so many of us [to buy] second or third homes, and homes before construction began, which we then sold to someone else who dreamed of riches from owning homes long enough to sell to another fool.

With the Fed’s easy money policy fueling the bubble, ARMs became a fairly safe tool. This is how I bought my first house, and two side-notes are relevant here. First, this was in the ’90s; ARMS weren’t just suddenly created in the 2000s, their use just expanded dramatically. Second, I can’t approach this as a morality play about subprime borrowers without indicting myself, so please understand I am not criticizing low-income home-owners (in fact I am far more critical of middle-class home owners who took advantage of subprime loans to buy big-ass houses than of lower-income home owners who jumped at the chance to strive for ownership of a modest home (our first home was 900 square ft[/efn_note].

Let’s say you are buying a $100,000 house, but don’t qualify for a 30 year fixed rate $100,000 mortgage. You can get a mortgage that has an introductory low rate that jacks up (sometimes in one year, but more often in 3 or 4) to a ridiculous amount. The initial low rate is to make it possible for you to afford payments, while the jacked up rate is to force you to refinance, because there’s no way in hell the bank can make money by giving you such a low rate permanently. It sounds nasty, but when the housing market is booming it works brilliantly, because you now owe, let’s say, $95,000 on a house that’s now increased in value to, say, $115,000. Sure a lender will give you a 30 year fixed rate mortgage for less than the house is worth; you’re a pretty good risk at that point because if you don’t pay they can foreclose and sell the house for more than they put out. The key is that housing prices have to keep rising, otherwise your interest rate jumps up and you’re stuck either paying the higher rate or defaulting because your house hasn’t increased enough in value to make giving you a regular 30 year fixed rate mortgage a good bet for the lender. This is particularly true if housing prices decline, because nobody’s going to give you a $95,000 mortgage on a house that’s only worth $92,000.

ARMs were so attractive as home prices boomed that by the time of the crash about 90% of subprime mortgages were of the ARM type. And the vast majority of defaults were by homeowners with ARMs.The default rate for other types of subprime loans wasn’t much higher than the rate for prime loans. In fact what is particularly notable is that ARMs were available for prime loans, too, and according to economist Stan Liebowitz, in 2006 and early 2006 both subprime and prime ARMS “sustained the same percentage
increase of foreclosures and at the same time.” This is more evidence that CRA by itself is not the primary cause of the mortgage crisis (although it is not wholly blameless, either).

These increasing prices were primarily a function of the interaction of the home-seller capital gains tax cut and the Fed’s easy money/low interest rates policy. And at that point just about every financial institution in the country wanted to jump in on it, and the number of ARMs boomed.

I want to pause to emphasize this point. This is the part most critics of the lenders looks at, but almost none of them looks at what led up to that point. They don’t pause to say, “what changed that caused the lenders to change their behavior and go from being stingy with ARMs to handy them out like cheap candy?” Something had to change at that point, and the most significant change was the reduction in interest rates. The capital gains tax cut itself was not sufficient to cause such a boom, but in conjunction with low interest rates the housing market caught fire, and a hot housing market justifies the use of ARMS. Yes, the lenders went nuts at this point. Yes, those who bought the securities backed by those mortgages went nuts. Yes, everybody should have known that the party couldn’t last. These financial institutions are in no way guiltless. But government policies create incentives and businesses respond to incentives. The policies were each well-intended, but good intentions had no relevance to the incentives created by their interactions.

How Mortgage-Backed Securities Work
Many mortgage lenders sell the mortgages they originate, in what is called the secondary mortgage market. This is what gives them the cash flow to offer mortgages to other people. Some people object to this practice of selling mortgages (and it sucks for the borrower when it gets sold to a financial firm with no interest in customer service–which they can get away with because they don’t loan directly to the public, so they’re not worried about getting repeat customers or scaring away potential customers), but imagine you were looking for a mortgage and all the lenders in your town were tapped out (the mortgage payments coming in aren’t necessarily large enough, collectively, to make new loans). In that case you’d be happy if they sold a couple of mortgages so they could have the cash to lend you.

The firms that buy the mortgages are, as a group, called mortgage aggregators because they assemble a bunch of mortgages into a pool (more accurately, many bunches into many pools). The most famous wholly private firm that did this was Countryside (which went bust and was taken over by Bank of America, which is still struggling to settle claims against it). You’ve also probably heard of Fannie Mae, which is a government-sponsored enterprise (GSE, chartered by the Federal Government), and which literally created the secondary mortgage market. It was created during the Depression specifically for the purpose of buying mortgages from banks so they would have the cash available to make more home loans. In the 1950s it was partially privatized, then in the late’ 60s fully privatized (the mortgage crisis plunged it into bankruptcy and receivership, and it was delisted from the NYSE, but you can still buy its stock, currently going for a whopping 28 cents a share).

The mortgage aggregators then sell securities backed by the cash flow from the mortgage, hence the term “mortgage backed security,” or MBS. An MBS is essentially the right to receive the cash flow from the mortgage payments. The securities are sold on the open market to whomever wants to buy them, such as financial investment firms, and the cash from the sales allows the aggregators to turn around and buy more mortgages from the mortgage lenders, which, as we saw, enables them to make more loans. So lots and lots of mortgages are really financed, although indirectly, by the financial firms buying the securities.

As noted, what they are buying is the cash flow from mortgages (minus fees snaked out by the mortgage aggregators), so your monthly mortgage payments (if your mortgage is part of a securitized pool of mortgages) gets passed through to whomever purchased the security. MBS came in two types, ones backed by residential mortgages and therefore called residential mortgage backed securities, or RMBS, and those backed by mortgages for commercial properties (such as office buildings and multi-family properties), called CMBS. For the most part it was the RMBS that played a big role in the debt crisis, because even though they weren’t a new innovation the proportion of subprime residential mortgages that were ultimately financed through such securitization increased from 50% in 2001 to 81% in 2005.

So, if the economy hums along and home prices keep going up, MBS are a good investment because they are very likely to pay off. That is, people are likely to keep making their mortgage payments, so the investment firms are likely to keep receiving their share of those payments. That makes them attractive investments, not just for the greedy but for any investment firm that a) has a fiduciary duty to their customers (all of them) and b) wants to keep attracting customers by offering them returns at least as good as they can get elsewhere (again, all of them).

But here’s the problem. When a whole bunch of mortgages–thousands upon thousands–are pooled together, the purchaser of the security can’t evaluate each of them individually to make a really sound analysis of the pool’s overall value, so they’re bought as much or more on faith as on any actual knowledge about their value. The advantage of buying an MBS is that the risk of any one mortgage defaulting is mitigated by the number of mortgages in the pool–if a few go unpaid, that loss is overwhelmed by the overwhelming majority that do get paid. But that only works if the great majority of the mortgages are sound. If too many go into default, there’s not enough cash flow to cover the obligations of the security; the issuer of the security defaults, and the purchaser of the security loses their investment. But nobody can actually see into the mortgage pools, so nobody really knows how sound they are–how many of those mortgages will avoid default. So you rely on proxy measures, and when all around the investment world you see pools that keep paying, it’s natural to assume the pools your securities are backed by will keep paying off, too.

Then it all went south because too many mortgages went into default all at once. The housing bubble couldn’t keep going forever, particularly when the Fed tightened the money supply by raising interest rates. This created havoc with adjustable rate mortgages in particular, as I showed above, because you can’t refinance an ARM on a house that hasn’t increased in value. That’s when people lose their homes, and that’s when financial companies lose their portfolios.

III: Glass-Steagall
So would Glass-Steagall have prevented this from happening? Investment banker James Rickards says so.

It was Glass-Steagall that prevented the banks from using insured depositories to underwrite private securities and dump them on their own customers. This ability along with financing provided to all the other players was what kept the bubble-machine going for so long.

But is that really accurate? First, we should be aware that Glass-Steagall was not really repealed in 1999; only part of it was. But it was the part that set up a “firewall” between commercial banks and investment banks. According to Peter Wallison, it was

the sections of Glass-Steagall that prohibited insured banks from being affiliated with firms—commonly called investment banks—that engaged in underwriting and dealing in securities.

So far that sounds supportive of Rickards’ position, but it’s not really. Wallison goes on to say that

The portions of Glass-Steagall that remained in effect after 1999 prohibited insured banks from underwriting or dealing in securities.

If that’s correct, then Rickards is wrong that banks used insured depositories to underwrite securities; or if he is right it means they broke the law in doing so, not that deregulation allowed them to do so. Further, he says, both before and after 1999,

banks were permitted to trade (that is, buy and sell) bonds and other fixed-income securities for their own account.

So if this is correct, the partial Glass-Steagal repeal changed nothing about what insured banks could and could not do–both before and after they could not underwrite securities, but they could buy and sell them for their own accounts.

Yaron Brook and Don Watkins concur, claiming that,

As for the FDIC-insured commercial banks that ran into trouble, the record is also clear: what got them into trouble were not activities restricted by Glass-Steagall. Their problems arose from investments in residential mortgages and residential mortgage-backed securities—investments they had always been free to engage in.

Ok, let me be up-front here. Wallison works for the American Enterprise Institute and Brooks and Watkins are Ayn Rand Institute folks, not the best recommendations for the readers of the League (and I’m certainly not enamored of Randians myself). But they are not making ideological claims here; they are making factual claims. If they are factually wrong, then please refute them. Here’s the Wiki entry on Glass-Steagall; note particularly section 3 discussing the provisions separating commercial and investment banking and section 5.5 on the 1999 Gramm–Leach–Bliley Act that repealed parts of those provisions. I can’t find evidence that Wallison, Brook and Watkins are wrong.

Besides, look at the firms that were hard hit–they were mostly investment firms, not commercial banks. IAG, an insurance corporation. Bear Stearns, an investment bank. Merrill Lynch, an investment bank. Lehman Brothers, an investment bank. Fannie Mae and Freddie Mac, mortgage backers. Not a commercial bank in the bunch.

There were some commercial banks that took a blow, but their stories demonstrate the point. Bank of America stumbled because it bought subprime lender Countrywide Financial. Countrywide was not, I don’t believe, the type of financial institution that B of A would have been prevented from buying under the repealed Glass-Steagall rules (I’m open to correction there; the distinctions between some of these financial firms is technical and I don’t claim expertise. However Countrywide was going down anyway, and the merging with B of A had nothing to do with that.) The same with Wachovia, which bought Golden West, another mortgage firm with garbage scows worth of bad debts.

This doesn’t mean Glass-Steagall (partial) repeal didn’t have any effect. Professional money manager Barry Ritholz, while stating that the Glass-Steagall repeal was “not a cause,” argues that the repeal exacerbated the crisis by allowing banks to merge and grow too big. Smaller banks failing don’t have as much effect as big banks failing. Economist Mark Thoma seems to agree, arguing that,

whether the elimination of the Glass-Steagall act caused the present crisis is the wrong question to ask. To determine the value of reinstating a similar rule, the question is whether the elimination of the Glass-Steagall act made the system more vulnerable to crashes.

Brad DeLong–you know, that notoriously liberal stimulus-lovin’ libertarian-hatin’ Berkely economist–quotes Thoma approvingly, but also goes so far as to put in his headline, “No, the Repeal of Glass-Steagall Was Not a Major Cause of Our Lesser Depression.” And in what seems to be a contra to Ritholz (although he’s not addressing him directly, and I may be misunderstanding him), DeLong seems to argue that the linking of commercial and investment banks provided stability, not vulnerability.

Joe Stiglitz thinks that the investment bank culture took over the commercial banks. I think that merging investment and commercial banks gave investment banks more stability in their liabilities that allowed them to ride out the storm more easily.

Alan Blinder agrees with me, which makes me happy.

… In an article on the causes of the current crisis, Joseph Stiglitz wrote, “When repeal of Glass-Steagall brought investment and commercial banks together, the investment-bank culture came out on top.”

But others note that the pure investment banks, like Lehman Brothers, have been the greatest source of instability, while the banks with combined commercial and investment arms have fared the best. “Banks did terrible things, investment banks did terrible things, a big insurance company named A.I.G. did terrible things, but basically none of that was enabled by the repeal of Glass-Steagall,” said Alan Blinder, an economist at Princeton…

Finally, let me turn to the Dean of liberal economists, Nobel Prize winner Paul Krugman.

It’s true that Glass-Steagalll, a Great Depression–era law that forbade the mixing of securities trading and accepting FDIC-insured deposits under the same corporate roof, wouldn’t have prevented the 2008 implosion of Wall Street. Instead, it was extraordinarily high levels of leverage at investment banks like Lehman and Merrill Lynch, as well as the holding of huge portfolios of toxic subprime mortgages by deposit-taking banks like Bank of America, that were the fuel for the conflagration. But progressives were right to feel that Wall Street had been dangerously underregulated for too long and that the entire country was now paying the price.

No backing off of the progressive line there, and as usual no love for deregulation from Krugman. And yet he says bluntly that “Glass Steagall…wouldn’t have prevented the 2008 implosion of Wall Street.” In fact, according to economist Arnold Kling,

No one says that restoring the separation between commercial banking and investment banking would prevent future crises. Hardly anyone even suggests that it would be helpful.

Now, let’s contrast that with what Blaise wrote.

The public policy we’re concerned about here is the elimination of the bright line between Wall Street investment banks and the FDIC controlled, retail banking institutions. Everything else is immaterial. Had those mortgages been kept on the retail side, this wouldn’t have happened because the banks wouldn’t have been sufficiently well-funded with Wall Street money to make more mortgages. Again, this arises from your imperfect understanding of how commercial paper and mortgage underwriting works.

Is it plausible that DeLong, Blinder, Krugman and Kling, highly respected economists all, were Blaise’s targets when he wrote, “You just don’t understand this well enough. If you did, you’d be making different noises”? Is it possible that DeLong, Blinder, Krugman and Kling, all of them, are suffering from an “imperfect understanding of how commercial paper and mortgage underwriting works”?

It seems to me that when someone insists they have a special understanding that is beyond the understanding of others, they have a duty and obligation to back it up, rather than limiting themselves to cheap assertions and smack talk. Is it possible I’m wrong here? Sure, but if you want to tell me I don’t understand, do your research and bring your A game, because I’m tired of the bush league bulls**t.