Everyone has their favourite theory about the financial crisis. It is now ten years since Lehman Brothers failed and most interested parties have decided what they think. It was deregulation, they claim, or bankers' bonuses, low interest rates, US government housing policy or moral hazard. There is very little hard evidence that any of these popular theories actually caused the particular disaster we experienced in 2008. Yet we continue to overlook an alternative account of the crisis that does explain some of its mechanics.
This alternative account, which for shorthand could be dubbed "regulatory error", does not result in easy policy prescriptions, it is geeky and it ascribes an annoyingly large role to that underrated driver of human history: ignorance. These are all reasons why it has been mostly ignored. But ten years on, it is surely time to release ourselves from the politics of it all and consider 2008 as an intellectual conundrum.
This is what Jeffrey Friedman and Wladimir Kraus tried to do in their short book Engineering the Financial Crisis in 2011. Seven years on, its uncomfortable conclusions are no closer to being accepted or debated than they were then. It explains, carefully and calmly, how pre-2008 banking regulations are likely to have played a role in encouraging many of the behaviours and effects that are now often blamed on deregulation or greed. And it suggests, worryingly, that we have doubled down on exactly the principles as underlay this regulatory system.
Friedman and Kraus set out to solve two problems. The first is why there was such a huge demand among regular US banks for assets later found to be at the centre of the storm: complex, mortgage-backed securities. The second is how these dud assets, despite being small cogs in the wider system, wrought such mayhem.
To answer the first question, they explore the operation of the Basel I capital regime. Basel I is a system of rules about how much and what kind of capital reserves banks ought to hold as a proportion of their assets. Ever since the Thirties, banks have been required to keep some cash aside to stop them gambling recklessly with government-insured money. But from the Eighties to 2001, the regime underwent several reforms. The most significant were the introduction of "risk weightings".
The risk-weighting system established a hierarchy, based on their historical performance, of how much capital a bank must put aside for each category of asset.
Government bonds required no reserves, whereas business loans required 100 per cent of the regulatory minimum. In between were mortgages. In the early 2000s, updates to the rules further attached great importance to credit ratings. Triple-A rated assets began to incur a much lower capital cost than lower-rated assets.
Accordingly, the market responded by innovating. Banks began to bundle together mortgages into securities and structure them into tranches, in which the top tranche would always be paid out first if anything went wrong. Because mortgage default rates would have to reach historically unprecedented levels for them to lose money, these top tranches were confidently assigned Triple-A ratings.
Selling these securities to Fannie Mae and Freddie Mac, which then resold them, added an additional layer of "safety" as both agencies were implicitly backed by government. Duly, from 2001 to 2006, regular US banks loaded up on these assets to the tune of $US1.2 trillion ($1.7 trillion). In doing so, these banks were not seeking out excessive risk and returns. They were buying something deemed ultra-safe by ratings and regulators.
Even then, however, these assets still only amounted to 12 per cent of the balance sheets for regular US banks (that is, commercial and retail banks). Kraus and Friedman ask how the losses prompted such a contraction in lending that it triggered the worst recession in nearly a century. In retrospect, the extent of the panic wasn't justified. Despite the high mortgage default rates, there was easily enough capital in the system to absorb losses without quite such a catastrophic credit crunch.
Again, the regulations played their part by interacting in ways that no one had predicted. Aside from the Basel regime, US accounting law required banks to update the value of their assets constantly in line with market prices. The aim of "mark to market" accounting was to prevent banks' own rosy assessment of their assets from obscuring real risks. But when this rule interacted with legal capital minimums and a market beset by panic and uncertainty, it dramatically exaggerated the shock. Banks had to mark down assets to such a degree that their only logical course of action was to shut off lending to avoid falling below capital minimums, even though the banks themselves thought market prices were wrong (about which they were subsequently proved right).
The complex interaction of regulations, incentives and market movements all combined to produce a credit crunch of unimagined proportions. As Friedman and Kraus write: "The failure to anticipate the mortgage bubble was shared by banker and regulator alike." Regulations did not create the bubble. But they do seem to have played a major role in over-concentrating risk in supposedly safe assets and then prompting markets to overreact.
Yet instead of rethinking the entire way in which we regulate banks, governments have decided to double down on the Basel regime. We have moved seamlessly from Basel I to Basel III and Basel IV. Each regime strips more and more discretion out of the system in search of "real" safety.
At the same time, thousands of pages of other regulations have upended the way derivatives markets work, overhauled rules for consumer lending, regulated shadow banking and so on. Some of these rule changes might have a positive effect. But in truth, we have no idea what their net effect will be and certainly no idea of how and when they will all interact.
The only certainty we can bank on, in fact, is that we are radically ignorant of how and when "the next crisis" will arrive. Just as bankers were ignorant of the risks they were taking and regulators ignorant of the risks they created, we're completely in the dark about the future. What 2008 should teach us, if anything, is that we could all show a little more humility.