Credit Cycles (Journal of Political Economy, Vol. 105, No. 2, April 1997)

"I mentioned, when discussing the Holmstrom-Tirole paper, that government may have to step in when the financial system suddenly loses liquidity. Now, how can that sudden loss of liquidity occur? Well, the Kiyotaki and Moore paper, Credit Cycles , provides a possible scenario. Typically, if a firm or bank borrows money for a project, it will have to provide some security – some collateral – for that loan. The Kiyotaki-Moore model assumes, with some justification, that borrowers can be induced to repay loans only if those loans are backed up by collateral of at least equal value: only the threat of losing the collateral provides the borrower with the incentive to repay. Get the weekly Five Books newsletter The project itself can constitute the collateral, but so can real estate or any other asset. Let’s take the case of real estate and imagine that its market value goes down a bit – perhaps because demand for housing turns out not to be as strong as people thought it would be. This decline in value can have a domino effect. Because the real estate is worth less, less lending is possible – since lending needs to be backed up by an equivalent amount of collateral. But if fewer loans can be made, there’s less production. And if there’s less production then people are poorer and therefore buy less real estate, causing a further fall in real estate prices, and you have a downward spiral. Like the others we’ve discussed, this model says a lot about the present financial crisis. What began as a problem in the real estate market – in fact, just the subprime part of the real estate market – grew through something like the Kiyotaki-Moore domino mechanism to envelop the entire financial sector, and, because the financial sector is critical for most production, to the real economy as well"