No, DeFi Is Not a Repeat of the 2008 Crisis

Charlie Warzel’s generally quite readable Galaxy Brain newsletter carried a provocative headline this week: “Is Crypto Re-Creating the 2008 Financial Crisis?”

Not surprisingly, it turned out to be a rhetorical question. The Atlantic writer’s newsletter carried an interview with American University law professor Hilary J. Allen in which she discussed her recent paper arguing that decentralized finance is repeating the mistakes of “shadow banking” that preceded the financial turmoil of the late 2000s.

You’re reading Money Reimagined, a weekly look at the technological, economic and social events and trends that are redefining our relationship with money and transforming the global financial system. Subscribe to get the full newsletter here.

Allen’s thesis is that the high degree of complexity around DeFi’s innovative new models for borrowing, lending, insurance and payments will leave the same lack of clarity around looming risks that credit default swaps (CDS) and collateralized debt obligations (CDOs) fostered during the pre-crisis housing bubble. “Complexity-induced opacity increases the chance that such risks will be underestimated in good times (causing bubbles), and overestimated in bad times (making panics worse),” she writes.

Allen is urging the U.S. government to step in to regulate the sector before it becomes more integrated into the mainstream financial system. She argues that decentralized applications (dapps) should be licensed and their founders and developers subject to enforcement actions if they are non-compliant.

That won’t sit well with many in the crypto community, where the idea that open-source coders can be charged with wrongdoing is seen as chilling to innovation.

First, let me acknowledge there’s some truth in Allen’s DeFi observations and that some of the parallels she draws to the financial crisis are legitimate and important.

It’s true the average person can’t hope to understand DeFi. Much like how Wall Street’s financial engineers exploited the black box of CDS and CDOs to the eventual detriment of bank customers, that complexity also gives DeFi project founders asymmetric advantages. It’s why “rug pulls” and other abuses of overly trusting investors are common.

Other valid observations from Allen: There’s an awful lot of 2008 bubble-like behavior in DeFi now, and there’s a lot more centralization with trusted intermediaries than “decentralization” enthusiasts acknowledge.

But there's a fundamental flaw in Allen’s perspective, one that could lead to a major policy error.

The elephant in the room

The overwhelming difference between DeFI innovators in the 2020s and those of Wall Street in the 2000s is that the latter – the bankers – operated within an all-encompassing political framework that the former – the crypto developers – are untouched by. With their power to create money through fractional reserve lending, banks function as the government’s agents of monetary policy, a specially sanctioned position that comes with privileged access to the Federal Reserve’s liquidity. There’s an interdependence between governments and banks that has, at times, morphed into codependence.

Exhibit A: the “too big to fail” problem in the lead-up to the financial crisis. This was the idea that a potential collapse in a big, systemically interconnected bank would pose such a catastrophic threat to the economy that the government would always have no option but to bail out such institutions if they ran into trouble – precisely what happened in 2008.

It was a moral hazard problem that, in the 2000s, fueled a massive market distortion. Before the crisis, banks faced asymmetric risks. They could profit from successes while the mortgage market was hot but faced no consequences if and when it turned south. The result was a warped, distorted version of capitalism in which profits were privatized and losses socialized.

In the reference that Allen makes to this, she mostly uses it to dismiss crypto enthusiasts as naive, suggesting their interest in DeFi is motivated by a disdain for bailouts. In reality, the federal government’s actions to shore up the financial system in 2008 were necessary. I think this completely misses the point. One can believe, as I do, that the 2008 bailouts were the lesser of two evils but at the same time criticize the “too big to fail” system of dependency that left the government no option but to execute them.

And that’s what’s hopeful about crypto. We have the prospect of freeing our financial system from dependency on the overly powerful intermediaries that have for too long commandeered an excessive proportion of the economy’s resources and political capital.

To achieve that, we don’t necessarily need to attain some utopian standard of total decentralization. (I find the “gotcha” critiques from the likes of Allen about how crypto’s not as decentralized as the narrative suggests rather tiresome. All the smart people in this space know this.) Rather, we need a system that is sufficiently open to competition and innovation for a significantly wider set of participants than exists in the current system. That means certain elements should be decentralized and “permissionless,” while other parts will require the involvement of trusted parties to achieve appropriate efficiency. What matters is the balance such that every institution is subject to some form of market pressure.

Easy innovation versus hard innovation

And that’s what makes the innovation-by-complexity comparison invalid across both realms. Since banks have a licensed monopoly over monetary creation, a role so vital that it earns them implicit taxpayer protection against losses, the “innovation” they undertake is shaped by very different incentives and checks and balances than that of DeFi developers. Banks had the luxury to develop CDS, CDOs and CDO-squared products to boost leverage and maximize short-term profits without having to calibrate those bets for the risk that the market might turn against them.

By contrast, DeFi developers face a much more fluid and unforgiving market. That’s not only because they don’t have the implicit taxpayer guarantee that banks have but also because of a core design element of DeFi: the open-source “lego” composability of code and low barriers to entry. That design means anyone with sufficient coding knowledge can spin up a new automated market maker, a new governance token or a new stablecoin algorithm without having to ask permission from a government or any other intermediating institution. And that means they can challenge the incumbents.

Consider the story of DeFi over the past two years. First, MakerDAO was the darling of the market, then Compound, then Aave, then Sushi Swap and then hybrid gaming/DeFi services like Axie Infinity, all founded within months of their sudden surge to success. Compare that with the winners that emerged from the rubble of the mortgage crisis: JPMorgan Chase and Bank of America. They trace their roots back to 1799 and 1904, respectively.

This DeFi dynamism, if it can be sustained, will prevent the rigidities that Allen worries will breed the same kind of systemic risk that consumed the banking system in the 2000s. That’s because the market is constantly correcting the different winners’ and losers’ tokens. It’s all about the price signals.

Also, while it’s true that DeFi is not perfectly decentralized and that it’s too complicated for the average person, end users of DeFi products have far greater influence over what gets built than do banks’ customers. Not only do many of them hold governance tokens, but with their fickle behavior they produce market signals that keep DeFi developers on their toes, something bankers don’t have to worry about nearly as much.

For sure, risk-taking investors will continue to lose money from rug pulls and code breaches while others will make fortunes. But this hurly-burly is quite different from the systemic problems that beset the financial system in the 2000s, when everyone and every risk asset was winning for a sustained period of years before everyone and everything started massively losing in unison. Most importantly, the constant threat of failure means there’s an incentive for developers to come up with more trustworthy offerings, which is why, despite the horror stories, the system has steadily become more robust over time.

What might threaten this market-driven balance? An ill-thought-out regulatory model. That’s what.

Want to build up systemic risk in DeFi? Then give banks, with their moral hazard-based lending model, an advantage over open-source developers. Make the latter seek permission to obtain the licenses that banks are already privileged to have. Make it very costly for real, market-focused innovation to occur and make short-term exploitative innovation virtually riskless by backing it with government insurance and taxpayer guarantees.

This is not to say centralized service providers in this space shouldn’t be held accountable to laws that preserve financial stability and protect consumers. But as a range of competing proposals for regulating stablecoins, DeFi and the broad crypto industry do battle in Washington, it does mean that we should heed the lessons from the 2008 crisis – the right lessons, not the wrong ones.