Susan Wharton Gates, Days of Slaughter: Inside the Fall of Freddie Mac and Why It Could Happen Again. p.4.
... the lack of contrition on the part of the former GSE regulator and top executives is shocking. Nor has there been any real remorse from the big banks, the brokers, consumer advocates, homebuilders, or real estate agents. In fact, nobody in the housing finance industry wants to talk about the past. But they have a lot to say about the future of housing finance: keep it pretty much the same. That is why I wrote this book—to tell an ugly story that everyone would like to forget.
—Susan Wharton Gates1
Freddie Mac is the large government-sponsored enterprise (GSE) that, along with its big sister Fannie Mae, dominated the market in housing finance in the years leading up to the financial crisis of 2008. Susan Gates worked at Freddie Mac from 1990 to 2009.
I also worked there, starting a few years before Gates and leaving over a decade before she left. I have only a vague memory of Gates, and I do not recall whether we ever worked in the same department at the same time. I do not appear as a "character" in her story. In two places in the book, she quotes from an essay published in 2010 in which I argued that the financial crisis could be viewed as a cognitive failure rather than a moral failure.2 As it turns out, Gates takes the view that the crisis was in large part a moral failure.
The fall of Freddie Mac came as a shock to those of us who were there in the late 1980s and 1990s, who refer to ourselves as "old Freddie." Was the tree that seemed so sturdy twenty-five years ago knocked down by a storm or did it rot from within? Gates says that it was both. She tells the story of Freddie Mac's fall as a combination of both external pressure and internal rot.
The external pressures came from Freddie Mac's peculiar quasi-public, quasi-private status. It pursued profits for shareholders (and bonuses for executives) while enjoying a number of advantages supplied by government, including a widespread perception among investors that in a crisis the government would back the debt of Freddie Mac (a perception that proved accurate). In describing these external forces, Gates provides an insightful description of the context in which Freddie Mac operated, starting with the political worship of home ownership that persists in the United States to this day.
By over-exalting home ownership, we remain fiercely wedded to out-dated exemptions that distort consumer behavior, encourage consumption, favor the wealthy, and reward increased indebtedness—all while contributing to budget deficits with little gain in homeownership rates. (Gates, page 47)
She accurately portrays the dysfunctional policies advocated by rapacious industry trade groups, self-styled "consumer advocates," and politicians of all persuasions. From a public policy perspective, it is an ugly picture, but one that you should study, no matter how much you might prefer to avert your eyes. Yet as important as the external forces were, I will devote most of the rest of this essay to the internal dynamics.
Both shareholders and policy makers pushed Freddie Mac to purchase riskier mortgages, the shareholders believing that such loans were profitable and policy makers believing that such loans benefited otherwise under-served borrowers, so they were in the public interest. In the end, the riskier loans proved to be unprofitable for Freddie Mac, harmful to the borrowers, and a disaster for the general public.
The irony is that "old Freddie" knew better. Long-time employees had spent decades resisting pressures to get involved in risky mortgage lending, but these employees lost out in the internal battles that took place, particularly in the fateful period from 2004 through 2007.
For more on these topics, see Freddie Mac: My Chapter, by Arnold Kling on EconLog. See also Financial Economics Crisis, an organized guide to Econlib links on the 2008 recession; Asset Backed Securities, by Philip Zweig in the Concise Encyclopedia of Economics, and the EconTalk podcast episode Kling on Freddie and Fannie and the Recent History of the U.S. Housing Market.
In Gates' telling, "old Freddie" is represented by David Andrukonis, with whom I worked closely. She describes his rise at the company as "meteoric," but to me it seemed like more of a steady climb. The "Dave A" that I know approaches business and life in general with a very strong sense of right and wrong. He is skeptical that a commitment to ethical behavior can be taught to adults, believing instead that either you have it or you don't.
Apart from Andrukonis, Gates does not introduce any other "old Freddie" characters into her story by name, although she includes several in the acknowledgments. On the other side, the "new Freddie" villains that are named include: Greg Parseghian, a Wall Street trader brought in to "grow the portfolio" in 1996; Mitch Delk, who headed the company's "Government Relations" department to which Gates transferred in 2001; and above all Richard Syron, who was installed as CEO by Freddie Mac's board of directors late in 2003.
Academic economists and others who have never worked in a large corporation can fail to appreciate the sort of internal conflicts that Gates describes. If you are used to thinking of a corporation as a unitary entity, the melodrama in her accounts of battles over memoranda may seem difficult to fathom.
Freddie Mac was two decades old when Gates joined, and her history of what preceded her is a bit sketchy. For example, she writes,
In 1970, Congress created Freddie Mac to securitize mortgages originated by S&Ls—and to give Fannie Mae some competition.
As I understand it, Freddie Mac was created by government to work around a problem that was caused by government in the first place. Savings and loan associations (S&Ls), which were the main source of mortgage lending in that era, faced ceilings (known as regulation Q) on deposit interest rates. As you would expect with price controls, this produced a shortage. The problem was particularly acute in fast-growing California. At that time, interstate banking was not allowed, so the S&L industry was segmented geographically, with no way to shift savings from the East Coast to the West.
Freddie Mac was created as a subsidiary of the Federal Home Loan Bank Board, and it was attached to the savings and loan industry. Freddie's role was to pool mortgages originated by S&Ls into securities and then to sell these securities to institutional investors, which often were S&Ls in other parts of the country.
The regulatory morass that provided a rationale for Freddie Mac gave rise to even more problems for the industry. When inflation and interest rates shot up in the 1970s, many S&Ls went bankrupt. However, because of "book value accounting," they were able to keep operating, using deposits guaranteed by the Federal Savings and Loan Insurance Corporation (FLSIC).
Suppose that an S&L had originated $10 million in mortgages at an interest rate of 5 percent. Subsequently, with market rates near 10 percent, those mortgages might have a current market value of $6 million. Thanks to book-value accounting, the S&L maintained the fiction that its mortgages were still worth $10 million, but it needed to avoid having to sell them, and meanwhile its deposits were melting away.
In the early 1980s, with FSLIC in trouble, Congress and regulators allowed Freddie Mac to "help" floundering S&Ls by letting them exchange old mortgages for securities without recognizing the losses on those mortgages. Under the Freddie Mac "Guarantor" program, the S&L with its book value of $10 million in mortgages could exchange those mortgages for securities that were valued at $6 million (paying Freddie a stiff fee for the privilege). Using those securities as collateral, the S&L could borrow funds in order to continue operating. You would think that at this point the S&L would have to record a loss of $4 million, because it had just exchanged mortgage loans with a book value of $10 million for securities valued at $6 million. But instead the Federal Home Loan Bank Board and other regulators deliberately shut their eyes and encouraged the S&Ls to treat this as an exchange of $10 million for $10 million. The regulators were just as eager as the S&Ls to "extend and pretend" that the institutions were still solvent.
As events proceeded in the 1980s, the strategy of "extend and pretend" lost viability. The government had to clean up a big mess, and taxpayers wound up with close to $150 billion in losses. It was at the end of the 1980s, having feasted on the carcasses of the S&Ls, that Freddie Mac asked for and received the privilege of getting out from under the umbrella of the Bank Board and to instead operate as a private company with shares traded on the stock market and available to the general public.
(Meanwhile, in the early 1980s, Fannie Mae was effectively bankrupt. Like the S&Ls, Fannie Mae had funded mortgage purchases with short-term debt. However, "extend and pretend" worked with Fannie Mae, in part because Fannie Mae changed its business model to reduce its risk, and in part because interest rates came down in time to save Fannie.)
I spent part of my first year at Freddie Mac on loan to Lawrence White, one of the members of the Federal Home Loan Bank Board. White fiercely opposed the accounting gimmicks that were used to keep defunct S&Ls in business. He had me write a paper (which was ordered to be published under a Freddie Mac executive's name) on the virtues of market-value accounting, which would allow regulators to see more clearly the true state of a financial institution.
By the 1990s, market-value accounting came to be the standard preferred by regulators. Ironically, it was blamed in part for the financial crisis of 2008. Critics argue that because institutions were forced to mark their holdings to market, as asset prices fell they were forced to raise more capital, which forced them to sell more assets, which drove down prices further, in a vicious cycle that caused asset prices to fall below where fundamentals should have taken them. Whether this actually happened, and to what extent, is something I cannot assess.
The employees that we think of as "Old Freddie" joined the firm either before or shortly after its shares became publicly traded. They had a deep knowledge of how the mortgage securitization process worked. RVO (since Gates refrains from using his name, I will do the same), who was Freddie's chief economist for many years, taught a lot of us the financial intricacies of mortgage prepayment and default, as well as the significance of capital regulation in shaping the competitive landscape in mortgage lending.
One of the lessons that we needed to learn the hard way came from the market for mortgages on rental properties. As of 1988, I was in charge of putting together models to price our guarantee for those loans. The department in charge of buying multifamily loans hated our models, which called for Freddie Mac to charge high fees for loans that it wanted to buy. They wanted lower fees in order to be able to book more business.
David Glenn, a new executive at Freddie Mac, brought this issue up late during an "operational planning" meeting with our department. He asked what we thought was our margin of error in our pricing models.
"Factor of 2," I blurted out. When I explained to him that this meant that when our model proposed a 40 basis point fee the range of reasonable numbers was 20 to 80, he was clearly not amused. He sent the rest of us out of the room in order to speak privately with my boss, PLD. When we came in the next morning and found that PLD and the rest of us still had our jobs, we experienced surprise and, of course, relief.
Within a few months, we would report that the rate of defaults on multifamily mortgages was coming in far higher than our models projected, which meant that, if anything, we were under-charging the originators who sold us those loans. It was the multifamily executives who got the axe, and subsequently all sorts of problems were discovered with the way that they had been conducting business.
We learned from the multifamily episode that putting a guarantee on mortgage loans can be a fragile business. If the default rate on loans exceeds one or two percent, the losses can more than wipe out the fees you earn on performing loans. Fortunately, our single-family guarantee business, which was profitable, was much larger, so that the losses on multifamily could be absorbed without bankrupting the company.
"Old Freddie" understood that as the S&L industry faded away, mortgage origination came to be dominated by unscrupulous, commission-driven salesmen. In order to protect itself from buying bad loans, Freddie Mac developed a protective shield consisting of underwriting requirements, compliance audits, and financial carrots and sticks to wield with mortgage originators.
On the other side of the market, Freddie Mac was selling mortgage-backed securities to Wall Street investors who were hardly less ruthless than the originators. "Old Freddie" understood this aspect of the business well, also.
When I first encountered David Andrukonis, his particular expertise was on "payment delay," a problem with Freddie Mac securities that was thought to be the source of their lower price relative to comparable securities issued by Fannie Mae. In order to address "payment delay" and other concerns, Freddie Mac launched an expensive initiative to replace its basic mortgage security with a new one. Dave, PLD, and I were the primary drivers of the initiative. The main challenge came from the information systems area. Those of us on the business side had a very naïve theory of how computer systems function in complex organizations, and by the same token, many staff in the information systems area had a rather naïve view of the structure of the business. This was a source of friction throughout my career at Freddie, and it was only years later, when I was working on my own business, that I came to appreciate the lessons that information systems executives like MGC and LNO tried to impart to me. By 1998, I was telling friends that my karma was that having spent much of my earlier incarnation at Freddie Mac berating information systems bureaucrats, in my new incarnation as an entrepreneur I had come back as one.
It was just at the point that the new security design was launched that Susan Gates enters the story.
I had been at Freddie Mac all of two weeks, and already a major celebration was underway. Employees were invited to toast the launch of Freddie Mac's newest securities innovation—the Gold Participation Certificate, or Gold PC for short. (Page 52)
As Gates points out, the new security failed to make Freddie Mac securities more competitive with Fannie Mae's. Perhaps the upgrade to Freddie's processing systems made the effort worthwhile, but that is hard to know.
"Old Freddie" also had beliefs about its proper role in the mortgage market. Financial executives wanted to have a AAA rating that was deserved, not simply granted by the backing of government. The internal principle was that the company should have enough capital to withstand a hypothetical downturn in the housing market in which average prices declined by 10 percent per year for four years and remained stable thereafter. This was known as the "Moody's Scenario," because it was that bond rating agency's portrait of another hypothetical Great Depression.
The only way to survive in the Moody's Scenario without an enormous capital cushion was to have a book of business of low-risk loans, particularly loans with initial loan-to-value ratios of 80 percent or less. That is, if a home was worth $100,000, the mortgage amount should be no higher than $80,000.
Had Freddie Mac stuck with the principle behind the Moody's Scenario, then it would have survived the housing price bust of 2007-2008. In fact, if it had stuck with the principle by maintaining its lending standards, then perhaps the bust would never have been as severe in the first place, because tighter lending standards would have taken away Freddie's contribution to inflating the housing bubble, and perhaps caused Fannie Mae and other participants to have second thoughts as well.
Gates implies that Freddie Mac's slide into impropriety and eventual self-destruction began with its purchase of private-label securities (PLS). The first mortgage-backed securities were all guaranteed by government agencies, such as Ginnie Mae or Freddie Mac. Securities that were issued by Wall Street firms without a government guarantee were called PLS. They faced considerable resistance from investors.
One of Greg Parseghian's innovations at Freddie Mac was to purchase PLS. In the short run, this allowed him to meet corporate objectives for increasing the size of the portfolio and earnings. In the long run, propping up the PLS market had consequences for Freddie Mac that were adverse. Growth in PLS enabled Wall Street to purchase more and more risky mortgages, helping to fuel the housing bubble and also adding pressure on Freddie Mac to lower its standards in order to maintain its status. Policy makers saw Wall Street catering to low-income borrowers and wondered why, with its government backing, Freddie Mac could not do the same.
Mitch Delk, an aggressive lobbyist, helped Freddie Mac to fend off various Congressional initiatives during the Bush Administration. One of Freddie Mac's tactics was to hire a slick public relations firm, the DCI Group.
Since it was abundantly clear that no one would listen to—much less believe—anything Freddie said or wrote, we decided we needed help: someone else to say what we wanted to say. It was called third-party advocacy, where a PR firm finds people sympathetic to your cause and assists them by writing an op-ed or speaking publicly on your behalf. (Page 195)
At that time, having left Freddie Mac a decade earlier, I was a regular columnist for a DCI-funded web site, called TechCentralStation. A Freddie executive (not Mitch Delk) let me know that Freddie Mac was willing to pay me to write a column about the virtues of GSEs. I refused. All of my columns were my own initiative, and I was determined to keep it that way.3
Delk was trying to maintain Freddie Mac's standing in the wake of an accounting scandal that came to be blamed largely on David Glenn, the company's number two executive. My own feelings about Glenn are mixed. I found him enigmatic and not easy to approach. Among employees, the rumor was that he and the CEO, Leland Brendsel, were on bad terms with one another. Brendsel's management style was very informal and instinctive. In contrast, Glenn seemed to operate by applying what he could learn from management consultants and business books. Known for his dedication to the Church of Latter Day Saints, Glenn by reputation had a plain and simple personal life. Brendsel by reputation did not.
Anthropologist Robin Dunbar famously suggested that humans can interact personally in group sizes no larger than about 150. In the late 1980s, Freddie Mac went from being what I call sub-Dunbar to super-Dunbar. When an organization is sub-Dunbar, it can operate informally. If I want to make a decision, I can call all of the people who might be affected. When an organization is super-Dunbar, I no longer know everyone who is going to be affected by my decision. A super-Dunbar organization needs formal procedures or else suffers from widespread miscommunication and employees working at cross purposes. Brendsel's management style was best suited to a sub-Dunbar organization. David Glenn took on the challenge of handling Freddie Mac's transition to super-Dunbar. However appropriate his management initiatives were to that task, they were not always well received.
As my experience with the multifamily episode illustrates, Glenn had a knack for making others feel uncomfortable. He came across as if he were waiting for others to trip up so that he could pounce on them. I thought that he failed to appreciate two popular, talented executives who were among my favorite co-workers. After MGC left the company, in large part because she could not handle working with Glenn, she was replaced by LNO, who also found David Glenn too difficult to work for and went elsewhere. Both of these highly capable women achieved better recognition at their subsequent firms.
Brendsel and Glenn both were ousted by Freddie Mac's board because of accounting scandals that came to a head in 2003. Parseghian was initially given the job of CEO, but he was not free of taint, and within two months the board also asked him to leave.
The accounting scandal had many odd aspects to it. For one thing, Freddie was accused of under-stating its earnings. The claim was that it was undertaking steps, such as beefing up loan loss reserves, which would reduce its earnings in the current quarter. Allegedly, the goal of this was to enable Freddie to show steady growth in subsequent quarters. Another charge was that Freddie was using "hedge accounting" for derivatives instead of marking them to market.4 This had been approved by Freddie's previous auditor Arthur Andersen, and subsequently the accounting profession has given its blessing to hedge accounting. But at the time, Arthur Andersen had to be dropped because of its role in the Enron scandal, and Freddie's new auditor did not approve of hedge accounting.
But one of the oddest aspects of the scandal was David Glenn's behavior.
Apparently, COO Glenn had not complied with a request [from law firm Baker Botts, acting on behalf of Freddie Mac's board] for his personal diaries. Later, when he turned over the diaries to Baker Botts, he indicated that they contained alterations and that pages were missing. (Page 180)
In the aftermath of the accounting scandal, Freddie Mac's Board found a new CEO, Richard Syron. As far as I can tell, no one from "old Freddie" has anything positive to say about the man, starting with his first speech to employees.
It probably wasn't the best thing to tell five thousand scandal-beleaguered employees that if they don't like working at Freddie Mac they should go work for the Department of Agriculture. He was trying to make the point that employees needed to toughen up. Still, the comment grated on everyone's nerves and was repeated for years to come. (Page 175)
"Old Freddie" took this as a sign of contempt. The feeling was mutual. As Gates puts it, Syron and his powerful chief of staff
... were smart men, but neither knew anything about the [intricacies] of running a mortgage shop. (Page 179)
The worst part is that they did not think that they had to know. They took it for granted that they could manage a push into high-risk mortgages without any understanding of what they were getting into.
The ultimate clash came in 2004, when Andrukonis, who was the company's Chief Risk Officer and at that time Gates's boss, sent an email to Syron, expressing opposition to loans where borrowers were not required to state, much less prove, that they had sufficient income or assets.
He sent an email directly to the CEO with copies to several executives, including me. In the now-famous email [it was discovered by Congressional staff and the media in 2008], Andrukonis told Syron flat out that NINAs [No Income, No Assets] were bad news for Freddie Mac:
Our presence in this market is inconsistent with a mission-centered company and creates too much reputation risk for the firm. An additional problem with this type of loan is that it appears they are disproportionately targeted towards Hispanics. The potential for the perception and reality of predatory lending with this product is great...
Exiting the NINA market would be difficult and expensive... since NINAs are minority rich, it will make it even more difficult to match the private market level of minority and underserved mortgage production. On the other hand, what better way to highlight our sense of mission than to walk away from a profitable business because it hurts the borrowers we are trying to serve... (Page 181)
Andrukonis had been making similar points in meetings for month. But for Syron and his henchmen, putting it in writing was embarrassing and unacceptable.
There was much table-pounding and red-faced expletives. Then came the veiled suggestion that the e-mail should somehow disappear. That was a red flag to a bull. Andrukonis was livid.
A few months later, another memo was slid into a file. At [the chief of staff's] behest, David Stevens, the senior vice president of sales, wrote a paean to the consumer benefits of [non-traditional mortgages]. It would be good for the history books to know there were two sides to the debate, right?
Although he would later be publicly exonerated, Andrukonis's career was over. Syron asked him to leave the firm...
Stevens went on to become the head of the FHA [Federal Housing Administration] and then the Mortgage Bankers Association... (Page 182)
Just to be clear: if during the run-up to the financial crisis, you were right about risky loans, you lost your job. If you were wrong about risky loans, you became an industry super-star.
Speaking of rewarding villains, Gates points out that in November of 2007, Freddie Mac disclosed that:
In addition to a $200,000 increase in base salary, Syron would receive "a special extension bonus" of $3.5 million, an additional equity grant of $800,000 in restricted stock, another equity grant for 2008 valued at $9.4 million, of which $8.8 million was guaranteed, plus a "special cash performance award" of up to $6 million, "depending on the number and strategic value of the performance milestones that have been achieved."
Employees were livid. The company's finances were in tatters.
In the wake of the housing finance disaster, policy makers and industry leaders have been quick to assure us that nothing like it will happen again. Based on her experience and analysis, Gates is not convinced.
Her book offers suggestions to policy makers to improve housing finance policy. But she concludes with a warning about the limits of formal rules. Right conduct, she argues,
... does not come from elected officials or regulators but from believing in something bigger than ourselves, something more moral than our hearts, an unseen judge of action and inaction who someday will call things into account. Only something like the old-fashioned fear of God has the power to induce people to do the right thing... (Page 253)
Susan Wharton Gates, Days of Slaughter: Inside the Fall of Freddie Mac and Why It Could Happen Again. p.4.
Arnold Kling, "The financial crisis: Moral failure or cognitive failure?". Harvard Journal of Law and Public Policy, March 2010.
Economists with more stellar reputations than mine were less reluctant to help the GSEs. See David Henderson's note on "Stiglitz and Orszags on Fannie Mae." EconLog, Nov. 8, 2009.
I am not an accountant, and I may be off base, but I understand it this way. Suppose that it is the month of May, and I am a farmer growing 100,000 bushels of wheat to harvest in August. To lock in the price, I sell a futures contract for August for 100,000 bushels for $10 million. Then suppose that in June the price of wheat drops 10 percent. The value of my wheat has fallen by $1 million, but the value of my futures contract hedge has gone up by $1 million. Under non-hedge accounting, I book the gain on my futures contract now, and I take the loss on my wheat in August, when I sell it. This makes my reported earnings volatile on a monthly basis. Under hedge accounting, I can wait until August to book my futures contract loss, so that my reported earnings are not volatile. Since the point of hedging is to stabilize economic value, as an economist I have a hard time seeing what the problem is (was) with hedge accounting.