Guillermo Calvo on the Crisis, Money, and Macro
What this covers
Guillermo Calvo and Russ Roberts examine the 2008 financial crisis through a lens that challenges mainstream macroeconomic theory. Calvo argues the crisis was fundamentally a liquidity collapse in finance rather than a failure of aggregate demand, and that the models economists relied on—particularly the IS-LM framework—had systematically obscured this reality. The conversation ranges across the theoretical and historical roots of that blindness, tracing how the discipline came to treat the financial sector as either self-equilibrating or irrelevant to monetary policy, an assumption the crisis violently disproved. Roberts guides Calvo through how liquidity as belief-dependent, how shadow banking created fragility, and what economic thinkers were actually paying attention to the problem all along.
Calvo traces the intellectual history deliberately: Irving Fisher's 1933 insight about debt deflation, Hyman P. Minsky's arguments on credit cycles, and the Austrian school's emphasis on boom-and-bust dynamics, all of which were sidelined as economics became attached to tractable models over realism. He locates the specific failure in 2008: shadow banks held mortgage-backed securities they believed were liquid, but when coordinated expectations shifted—triggered by subprime fears—a sudden stop in credit followed because the securitization channel collapsed. He criticizes both the Keynesian theory of crisis onset (which rests on animal spirits without mechanism) and the monetarist response (money-supply expansion alone), arguing instead that policy should address the credit channel directly. Calvo also considers how the Fed's deployment of Quantitative Easing and asset purchases reflected an implicit recognition that conventional tools had broken, and he examines how low interest rates in safe assets drive liquidity chasing into emerging markets, replicating the same fragility abroad.
Calvo argues that the 2008 crisis was fundamentally a liquidity crunch in the financial sector that triggered a credit 'sudden stop,' and that mainstream macroeconomics—by treating finance as a 'veil' and relying on simple models like IS-LM—failed to understand it.
- Liquidity is fragile and 'in the eye of the beholder,' so it can be destroyed by coordinated expectations without any change in real fundamentals
- Macroeconomists treated the financial sector as self-equilibrating and irrelevant to monetary policy, an assumption exposed as false in 2008
- Keynesian, monetarist, and modern calibrated models offer ex-post 'magic pills' (aggregate demand, money supply) without diagnosing the financial mechanism
Liquidity is fragile and non-fundamental; shifts in beliefs about asset safety cause sudden stops and runs.
- An asset's liquidity depends on other people believing it is liquid—like paper money, which is accepted only because the recipient expects someone else will accept it—so liquidity can vanish suddenly when beliefs shift, even though the asset's real characteristics are unchanged.
“the liquidity depends on other people thinking that instrument is liquid. It's like with paper money.”
- When a substantial number of market players coordinate expectations and doubt MBS quality (fearing subprime contagion), a run begins; because the underlying mortgages pay out over 15 years, there is not enough cash to maintain prices, so when many try to sell simultaneously buyers step back and prices collapse.
“Somebody says a substantive number of market players, for some reason, coordinate expectations--they start to doubt... Well, that's good enough to provoke what we call a 'run.'”
- Shadow banks held mortgage-backed bonds as assets and borrowed heavily against them rather than using equity; depositors felt safe—even without FDIC protection—because they believed the assets were liquid and could be sold instantly, a belief reinforced by the low volatility of the Great Moderation.
“the banks, what they did, shadow banks in particular, held those bonds in the assets and borrow against those bonds.”
- When confidence in shadow banks is shattered, savers stop depositing and instead hold cash or Treasury Bills, withdrawing money from the credit market so credit dries up immediately—a liquidity trap—and the surge in demand for safe assets pushes interest rates sharply down, including demand for dollars even though the crisis originated in the US.
“the depositor keeps the money in his pocket. Or tries to buy Treasury Bills. So you take the money away from the credit market, and that's how credit dries up immediately.”
- Unlike Keynesian investment contractions, which require a large change in investment to produce big output effects, liquidity is so non-fundamental that it can be destroyed by only a little 'animal spirits'—a small coordination shift—because there are multiple equilibria, whereas destroying real capital, productivity, or technology requires far more.
“When you think about liquidity, it's something that can be destroyed without any big switch anywhere. Because there are multiple equilibria--if I can use economists' jargon.”
Finance is interconnected and systemic, not separable from macro; credit crunches spread across the economy.
- The financial sector is like the nervous system—it touches every part of the economy through interconnected credit relationships (housing to construction to suppliers, plus inter-firm trade credit), so a credit crunch in one sector spreads as borrowers and lenders become uncertain about each other's solvency and grow risk-averse.
“The problem with the financial sector is like the nervous system. It touches every note of your body.”
- Macroeconomists could safely leave out the financial sector because they only consider a problem worthy of attention if it is macro or systemic; individual bank failures are micro problems, and the defining feature of the 2008 crisis is that the financial problem became systemic.
“a problem deserves attention if it is a macro problem or if it is a systemic problem”
- For the entire 20th century, mainstream macro models and central banks assumed a sharp separation between money and the financial sector—that finance would equilibrate itself without intervention and that monetary policy had essentially nothing to do with financial borrowing and lending—an assumption that proved untenable in 2008.
“the assumption is that there is a sharp difference between money and the financial sector. And that the financial sector will take care of itself.”
- The dot-com tech bubble had mild macroeconomic effects—firms went bankrupt and workers were laid off, but the broader economy was barely affected—which reassured macroeconomists that the financial sector could take care of itself, making 2008 a surprise.
“the tech bubble kind of reassured macroeconomists that there was no need for them to be paying too much attention to the financial sector”
QE targets credit channel repair, not money supply; fiscal stimulus alone cannot substitute for financial sector repair.
- The supposed proof that fiscal stimulus works comes from WWII, when government expenditure roughly doubled, but replicating that scale of spending today is politically unthinkable, undermining the claim that the Keynesian 'pill' has been properly tried.
“We tried that pill in the second World War, and I was looking at some numbers: the government deficit [expenditure] increased by, I don't know, 600%, something like that... expenditure, doubled on average during the war years.”
- The 'money is a veil' proposition—that increasing the money supply only raises prices proportionally without real effects—rests on special assumptions; once wage or price rigidities exist (as Keynes showed), money affects not only prices but also output.
“that result is based on some assumptions that, when you make them explicit, you realize that you are talking about a special case”
- QE is better understood as a 'defragmentation policy' aimed at reopening the broken credit channel rather than as conventional money-supply expansion; it should not be necessary to drive the policy interest rate all the way to zero before using QE, since the zero lower bound itself signals that the Fed's interest rate has very little impact on the private credit market.
“QE, I call it 'defragmentation policy.' It's a policy to try to open up, in my mind, to open up the credit channel.”
- The Fed went beyond pure monetarism: the first QE was an attempt to buy toxic assets and commercial paper—unprecedented heterodox actions—showing they recognized that simply increasing the money supply à la Friedman was insufficient, since otherwise there would be no reason to purchase toxic assets.
“the first QE was an attempt to buy toxic assets, indicates that they went beyond just controlling the money supply a la Friedman. This was a rather heterodox policy”
- Lowering interest rates on very liquid US assets incentivizes investors to search for yield in other liquid assets, encouraging flows into emerging markets; when those markets later feel a threat that US rates will rise or US liquidity will tighten, the funds fly back, producing in emerging markets the same liquidity crunch and sudden stop—as seen in the taper-tantrum reaction.
“by lowering the interest rate on very liquid assets you may be giving incentives for the system to... go out in search of other liquid assets... And that may have encouraged the flow of funds that went into emergent markets”
- Since the main factor behind the crisis lies in the financial sector, it is a strange leap of faith to conclude the solution is a large increase in government expenditure; policy should first focus much more on how to fix the financial sector itself rather than jumping to fiscal stimulus.
“It sounds very strange when you realize that the main factor behind the current crisis lies in the financial sector, to say that solution should then be an increase in government expenditure. There's a big jump there.”
Debt-deflation—falling prices raising real debt burden—triggers bankruptcies, liquidity traps, and persistent economic damage.
- Although the policy of keeping the CPI from falling succeeded, the CPI does not adequately capture housing prices, so the indebted homeowners who bought houses still face exactly the debt-deflation problem Fisher worried about—meaning aggregate price indices can mask sector-specific debt deflation.
“the CPI has not collapsed but those who are indebted, being indebted in order to buy a house, face exactly the same problem that he was concerned with--the debt deflation.”
- Irving Fisher's 1933 insight was that precipitous price declines (wholesale prices fell over 30% in the 1930s) raise the real burden of nominal, non-state-contingent debts—loans whose payments are fixed regardless of the borrower's circumstances—making some debts unpayable, triggering bankruptcies, flight to safety, and a liquidity trap.
“when prices fall precipitously like it happened in the 1930s, that it increases the real value of the debt and may make some debts not payable. And lead to bankruptcies”
- Friedman and Schwartz argued the Fed's mistake in the 1930s was failing to increase the money supply to keep prices from falling, but they treated this as an empirical observation without explaining why; they ignored Irving Fisher, who provided the rationale—falling prices raise the real value of debt and cause bankruptcies.
“He pushed very much the idea that the big mistake in the 1930s was for the Fed not to increase the money supply and keep prices from falling. He doesn't discuss why.”
Securitization transformed illiquid mortgages into traded bonds, enabling credit expansion that masked underlying fragility.
- Once the securitization channel collapses, lending must revert to direct loans to house buyers at higher interest rates; this jump in the cost of credit to borrowers is a 'sudden stop' in credit, equivalent to the abrupt halt in financing that emerging markets had experienced in earlier crises.
“when that channel sort of collapses momentarily, then you had to go back to lending directly to the house buyers. So the price, or the interest rate that the house buyer faces now increases. And that's equivalent to a sudden stop in credit.”
- Mortgage-backed securities transformed illiquid individual mortgages—which required knowing each borrower's characteristics to sell—into liquid, traded bonds, because pooling makes the income flow statistically predictable without knowing any individual borrower, creating an asset more attractive than money since it yields a positive return.
“mortgage-backed securities is a very clever device by which the banks put all of those mortgages and bonds, and the bond is traded in the capital market. So what that does is to make your mortgage more liquid.”
Keynes lacks a theory of crisis origins; Hayek and Mises understood credit booms foster busts but remain ignored.
- Keynesian theory applies to what to do after a crisis but has a very poor theory of how crises arise—the only mechanism Keynes offers is 'animal spirits,' which is essentially an admission of not knowing and fails to explain the coordination that causes everyone to suddenly judge investment a bad idea.
“Keynes does not have a good theory--in fact, his theory is very, very poor--about how the crisis comes about. The only thing he mentions in the book is animal spirits.”
- Mises and Hayek were very aware that credit booms could foster credit busts and forcefully made that case, offering insights highly relevant empirically today, but they are ignored in standard top US economics courses partly because their work is difficult to read and was never developed into a formal theory comparable to 20th-century mathematical economics.
“these guys were very well aware that these, for example, credit booms could foster a credit bust, and so on. And these two authors in particular, they were very forceful about making those cases.”
Academic economics favors simple tractable models over realistic ones; policymakers innovate better by facing real constraints.
- Top academic places are unlikely to generate useful fresh ideas, whereas central bankers and policymakers are more innovative because they must face the real world and cannot hide behind a model.
“policy makers are very active... Because those guys, they have to face the real world. They cannot pretend, I mean hide themselves, behind a model.”
- Simple models carry the day in economics—the IS-LM model was powerful precisely because it was simple, but it was too simple—and this dynamic explains why subtle thinkers like Minsky failed to persuade even sympathetic economists like Tobin, illustrating a systematic bias toward tractable rather than realistic models.
“the lesson is that simple models carry the day. And the danger is that, precisely--simple models carry the day. And so the IS-LM was so powerful, perhaps because it was very simple. But it was too simple.”
- Powerful inertia in academia—graduate students must follow advisers and publish in good journals, and thinking is costly—keeps producing theoretical NBER papers whose key assumption is a perfectly functioning credit market, then applying them to policy debates, when that very assumption assumes away the central problem of the crisis.
“graduate students being fettered by their advisers and following on those lines because the want to get a Ph.D. and then they want to get their papers published in good journals... So there is an inertia, an enormous inertia there.”
- Calibration—taking a purely theoretical model, feeding in shocks and adjusting them until the model replicates observed data—creates a false sense of validity, because enough tinkering will eventually hit the mark without bringing the model closer to reality, and it absorbs effort into programming rather than thinking about the real world.
“if you keep playing with the shocks, eventually you hit the mark. But that doesn't mean you are closer to reality.”