Why Things Break: The AAA Rating Was Correct

Published:

Why Things Break: The AAA Rating Was Correct

A house in a falling market. The guarantee was the scandal, not the rating.

The standard story of the 2008 financial crisis goes like this: greedy bankers bundled bad mortgages into securities, corrupt rating agencies stamped them AAA, and the whole pile collapsed when the lies came due.

Almost every element of that story is wrong. The interesting question is not why those securities got AAA ratings. The answer is obvious. They were backed by the US Government. The interesting question is why the government was guaranteeing them at all.

This is the first non-IT case study in Why Things Break, the series that asks of every disaster: who benefited, who paid, and what incentives produced the outcome. The introduction promised case studies of government IT failure - the NHS’s ten billion pound National Programme for IT and Birmingham City Council’s Oracle project. This piece takes the same question to a different kind of government failure: the deliberate manufacture of a housing bubble. Both follow the same pattern: people who do not understand the thing they are approving make the decisions, the people who could warn them are not in the room, and nobody faces consequences.

The housing bubble is the same disease, at a scale that makes Birmingham look like a rounding error. It crashed several economies at once. And the tell that reveals the whole mechanism is that AAA rating.

The story: how the government built a bubble

The root cause was not a tax cut. It was not deregulation. It was a government decision that home ownership is good, followed by a decades-long campaign to manufacture it by force-feeding credit to people who could not afford it.

Start with the causal logic, because everything follows from it. People who own homes tend to be happier, healthier, more productive, more likely to be married, less likely to be involved in crime. True enough. But the direction of causation runs from stability to home ownership, not the other way. People who are employed, sober, and settled buy houses. Houses do not make people employed, sober, and settled.

The policy class read the correlation backwards. They decided that if home ownership produces good citizens, then manufacturing home owners will manufacture good citizens. So they pushed home ownership directly, instead of the boring, slow, unglamorous conditions that actually produce it.

And they pushed it with a very specific tool: the government guarantee.

The machinery took decades to assemble. The Community Reinvestment Act, passed in 1977, pressured banks to lend in underserved neighborhoods. The 1992 law that created the affordable housing goals gave the Department of Housing and Urban Development the power to tell Fannie Mae and Freddie Mac how much of their business had to be loans to low- and moderate-income borrowers. Those goals were raised, and raised again, through the 1990s and 2000s.

The moment that matters is 1997. That is when First Union and Bear Stearns priced the first securitization of Community Reinvestment Act loans - $384.6 million of securities, guaranteed by Freddie Mac, carrying an implied AAA rating.

Read that sentence again. The federal government, through Freddie Mac, guaranteed private loans made to borrowers who could not qualify for normal mortgages, and turned them into AAA securities. Subprime lending had existed before 1997, but as a niche. The guarantee is what scaled it into an industry. Why would investors buy anything else? Here was a security that paid better than a Treasury bond and was backed by the same people. The AAA rating was not a fraud. It was an accurate description of a security backed by the American taxpayer.

The machine then ran on its own incentives. Fannie Mae, under pressure to meet the HUD goals, announced in 1999 that it would ease its credit requirements to make mortgages available to more low-income borrowers. Andrew Cuomo’s HUD kept raising the goals. Banks got fees for originating the loans. The GSEs got fees for guaranteeing them. The securities got AAA ratings because they were, in effect, government bonds with a better yield. Everyone in the chain was paid. The only person in the transaction who was not at the table was the one who would eventually pay for all of it: the taxpayer.

Add the Federal Reserve. From 2001 to 2004, Alan Greenspan’s Fed held interest rates at historic lows, then kept them at one percent through 2003 and 2004. Cheap money is the fuel; the guarantee is the engine. When the Fed makes borrowing nearly free and the government guarantees the riskiest loans, you do not get more responsible home ownership. You get a housing bubble.

The crash

The bubble peaked in 2006 and began to deflate. By September 2008 it had taken the financial system with it. Lehman Brothers failed. The government put Fannie Mae and Freddie Mac into conservatorship - the largest bailout in American history, roughly $187 billion of taxpayer money injected into two companies that had been running a private-profit, public-risk machine for a decade.

The damage was not confined to the United States. This is what separates the housing bubble from a normal policy mistake: it crashed several economies at once.

  • United Kingdom. Northern Rock, the first British bank run in 150 years, was nationalized in 2008. Royal Bank of Scotland received a £45 billion bailout, making the state its majority owner. The total UK banking rescue peaked at over £1 trillion in guarantees and liquidity support.
  • Ireland. The government guaranteed the liabilities of its banks in September 2008 - a decision that destroyed the state’s own credit. Property prices fell by roughly half. The bailout that followed, in 2010, was €85 billion, and the country spent a decade paying for it.
  • Spain. The construction boom collapsed, the savings banks known as cajas were wrecked, and the banking system needed a €100 billion European bailout. Unemployment hit 26 percent; youth unemployment passed 50 percent.
  • Iceland. All three major banks collapsed within a week in October 2008. GDP fell by roughly ten percent. The currency lost half its value.

Ireland and Spain built their own bubbles - cheap euro credit and construction-led growth, not American subprime. They did not need the US policy machine to ruin them; they had their own version of it. That is the point. The same incentive structure, wearing different national costumes, produced the same result on both sides of the Atlantic.

In the US alone, roughly ten million households lost their homes to foreclosure, and about seven trillion dollars of household wealth evaporated.

The test: who benefited, who paid?

Run the incentive trace, the way this series does with every disaster.

Who benefited? The politicians who announced affordable housing initiatives got ribbon-cuttings and a bipartisan glow. Everyone was for home ownership; nobody was for foreclosures. The banks got origination fees and the implicit backstop - private profits, and when it failed, public rescue. Fannie Mae and Freddie Mac executives got bonuses tied to growth in exactly the loan books that later collapsed. The ratings agencies got their fees and gave the ratings that the facts - the guarantee - supported.

Who paid? The taxpayer, for the bailouts and the trillions in lost wealth. The ten million families who lost their homes, many of them the very people the policy was supposed to help. The young, who carried the recession’s unemployment and debt into their thirties. The Irish, the Spanish, the Icelanders, the British - entire economies handed a bill for a decision made in Washington and mirrored in their own capitals.

Let me make this personal, because it is. In 2008 I lost my job. It had nothing to do with the crash - a company decision, a different decade’s problem. I was 53 years old. I told my wife we would have to downsize. She told me she wanted a divorce. So there I was: unemployed, getting divorced, and suddenly standing in a falling house market, trying to off-load my only real asset - a house bought at a price the government’s policies had helped inflate, in a market that was now falling around it. Thanks everyone.

And notice what did not happen. No politician who voted for the affordable housing goals lost their seat over the crash. No HUD secretary who raised the goals faced consequences. The executives took their bonuses; a few went to prison in Iceland, but the policy class that wrote the rules did not. As with the NHS and Birmingham, the decision-makers were insulated from the results of their decisions. The people who write the goals are never the people who pay for them.

The deeper point is the one this series keeps returning to: there were no technocrats in the room. The people who understood credit risk - the underwriters, the actuaries, the loan officers who had lived through the savings and loan crisis of the 1980s and knew exactly what unqualified borrowers do - were not the people setting the affordable housing goals at HUD. The people setting the goals understood politics, not credit. They saw a correlation, decided it was a cause, and built a machine that turned the mistake into a global disaster.

The pattern: it was not the first time, and it was not the last

The United States ran this exact play in the 1930s, when the Federal Housing Administration was created to push home ownership and ended up seeding a different set of distortions that lasted fifty years. It ran it again in the 1990s and 2000s with the CRA goals and the GSE guarantees. And it is running it again right now: first-time buyer subsidies, FHA loan limits raised, down payment assistance programs, all of them aimed at the same target - more home ownership, manufactured by making credit easier and riskier, with the taxpayer as the silent partner.

The United Kingdom ran its own version with Help to Buy, which pushed up prices and handed developers a state-subsidized sales channel. Ireland and Spain ran theirs with cheap euro credit. The pattern is always the same: announce a goal everyone agrees with, take the credit, leave before the bill arrives, and let the next government - or the next generation - pay it.

How many times does a policy have to fail before the people who propose it face a cost? The answer, so far, is: an unlimited number. Because the cost never lands on them.

The lesson: the AAA rating was the tell

The next time someone tells you the crisis was caused by corrupt rating agencies stamping AAA on junk, ask them a question: what was the security actually backed by? The answer is the US government, through Fannie Mae and Freddie Mac. The rating was correct. The securities were, in effect, government debt with a better interest rate. The ratings agencies were not the scandal. They were the honest bookkeepers of a dishonest arrangement.

The scandal was the guarantee. The government took private risk, put its own credit behind it, and gave the profits to the private sector while keeping the losses for itself. That is not a market failure. That is the state manufacturing a market for its own policy goal, and charging the citizen for the privilege.

This is the same lesson as the IT disasters, one level up. When the government decides that something is good - a national health record, a council’s finances, home ownership - and backs that decision with borrowed money and a guarantee, it removes the one mechanism that keeps private decisions honest: the risk of being wrong. In the private market, someone who guarantees bad loans loses money and stops. In the state, the people who guarantee bad loans get promoted.

The choice is not whether we have home ownership. It is whether home ownership is bought honestly, by people who can afford it, or manufactured with borrowed money and a government guarantee, and paid for later by everyone. We have seen what the second option does. It crashed several economies at once, and the people who built it are still in their jobs, ready to build the next one.