This fall I am taking a course on the “international financial crisis” taught by Jon Macey and Greg Fleming (yes, the former COO of Merrill Lynch). The first assigned reading is a speech that Larry Summers gave at the AEA in 2000 entitled “International Financial Crises: Causes, Prevention, and Cures,”* summarizing the state of the art in preventing and combating financial crises. It’s based on experiences from emerging market crises in the 1990s, and doesn’t even contain a hint that something similar might happen here; however, few people could fault Summers for making that oversight back in 2000, and I certainly won’t.
Many people, including Simon and me, have discussed the similarities between our recent financial crisis and the emerging market crises of the 1990s, so I’ll be brief. The main similarities are excessive optimism that creates an asset price bubble, a sudden collapse of confidence that causes the rapid withdrawal of money and credit, a liquidity crunch, and rapid de-leveraging that threatens solvency. (We have also argued that there are political similarities, but let’s leave that aside for now.) The biggest difference is that instead of being compounded by flight from the affected country’s currency and government debt, in our case the exact opposite happened; investors fled toward the U.S. dollar and Treasuries, making things easier for us than for, say, Thailand. Also, to a partial extent, the parallel requires an analogy between emerging market countries and United States banks; for example, the issue of bailouts and moral hazard arises in the context of the IMF bailing out Indonesia and in the context of the United States government bailing out Citigroup.
Summers’s speech makes a lot of sense, so I’ll just highlight a few points he makes that I think are particularly instructive given our recent experience. I think these are all excellent points. For each one, I’ll quote from Summers, and then comment on its relevance to our situation.
1. Financial crises result from fundamental problems that should be fixed.
“It seems difficult to point to any emerging-market economy that experienced a financial crisis but did not have significant fundamental weaknesses that called into question the sustainability of its policies.”
“Bank runs or their international analogues are not driven by sunspots: their likelihood is driven and determined by the extent of fundamental weaknesses. … Preventing crises is heavily an issue of avoiding situations where the bank-run psychology takes hold, and that will depend heavily on strengthening core institutions and other fundamentals.”
In other words, you can’t blame a crisis entirely on investor psychology; there is something rotten. The panic of September-March was not, as some argued, simply a liquidity crisis; the premise of the liquidity crisis theory is that the fundamentals are sound (institutions are solvent), and Summers doesn’t believe that.
2. The strength of domestic financial systems and institutions matters more than aggregates such as the amount of debt.
“When well-capitalized and supervised banks, effective corporate governance and bankruptcy codes, and credible means of contract enforcement, along with other elements of a strong financial system, are present, significant amounts of debt will be sustainable. In their absence, even very small amounts of debt can be problematic.”**
The fact that U.S. consumers and banks had a lot of debt isn’t itself a cause of anything; if our financial system were effectively governed, debt alone would not have brought it down as spectacularly as it did.
3. Short-term funding is dangerous because it can be difficult to roll over in a crisis.
“Policy biases toward short-term capital need to be avoided.”
“A measure of sound management of short-term flows is implicit in any prudential regulation of banks.”
The U.S. failed on this count. A large portion of the shadow banking system – SIVs and SPVs raising short-term funds and investing them in long-term assets – was entirely dependent on short-term capital. It also turned out that most of the large corporate sector was dependent on commercial paper, which is also short-term. When short-term funding vanished in September, everyone was stuck.
4. Transparency matters.
“If one were writing a history of the American capital market, I think one would conclude that the single most important innovation shaping that market was the idea of generally accepted accounting principles. The transparency implicit in the generally accepted accounting principles (GAAP) promotes efficient market responses to change, and it supports stability. Furthermore, if as Ken Galbraith has observed, conscience is the fear that someone may be watching, it may be the single most effective means of promoting self-regulation.”
I think Summers is right on a historical scale – standard accounting conventions promote transparency. But looking only at the last two decades, I thnk that GAAP did not keep up with the realities of financial innovation, for example in accounting for SIVs. For the beneficial effect cited by Summers to hold – accounting standardization promotes effective self-regulation – the accounting has to accurately reflect the true risk being taken by institutions. If not, the causal chain breaks down.
5. Expectations of bailouts are bad.
“While conditioned, precautionary financial support is constructive in some cases, the risk inherent in systematic availability of unconditional credit to countries can be summarized in two words: moral hazard.”
“It is certain that a healthy financial system cannot be built on the expectation of bailouts.”
6. Rapid intervention in unhealthy institutions is critical.
“Prompt action needs to be taken to maintain financial stability, by moving quickly to support healthy institutions and by intervening in unhealthy institutions. The loss of confidence in the financial system and episodes of bank panics were not caused by early and necessary interventions in insolvent institutions. Rather, these problems were exacerbated by (a) a delay in intervening to address the problem of mounting nonperforming loans; (b) implicit bailout guarantees that led to an attempt to “gamble for redemption”; (c) a system of implicit, rather than explicit and incentive-compatible, deposit guarantees at a time when there was not a credible amount of fiscal resources available to back such guarantees; and (d) political distortions and interferences in the way interventions were carried out.”
Again leaving aside (d), I think our government was guilty of (a), (b), and maybe (c) in the recent crisis. It’s not clear that (a) has yet been satisfactorily addressed (even PPIP seems to be evaporating), especially when it comes to commercial real estate, and we clearly have (b). As for (c), we did move to explicit deposit guarantees, but we still have implicit guarantees on other bank funding (bonds); and though the government probably has a credible amount of fiscal resources, there is still the question of whether Congress or the Fed will come up with the money in a pinch.
So again, I think the Summers of 2000 was basically spot-on. However, I think the Obama Administration – of which Summers, of course, is a central figure – is doing a spotty job of implementing his lessons. Most notably, we have increased the expectation of bailouts by supporting unhealthy institutions, increasing moral hazard; and we have left the problem of toxic assets largely untouched, hoping that economic recovery will make it go away. I think this has been due largely to political realities, not to any failing of Larry Summers to understand his own thinking; the United States government has a lot more power negotiating with itself than an emerging market country has negotiating with the IMF.
The big current question is whether financial regulatory reform will fix the underlying problems that, according to the Summers, are the root of financial crises. For example, according to Summers (2000), we want a regime that discourages dependence on short-term funding, we want more transparency, and we want something that reduces rather than increases expectations of bailouts. Whether we get that – or whether the administration prefers to let regulatory reform fade away in the wake of the health care war – is the remaining test.
* The offiical online home of that paper doesn’t allow free downloads, so I don’t think I should post my copy, although other people may have.
** Citing a paper by Simon, Peter, Alastar Breach, and Eric Friedman!!