In this post, we depart from our usual practice of addressing research by other authors and discuss a recent paper of our own: "Reasonable people did disagree: Optimism and pessimism about the U.S. housing market before the crash." We first review the main points of the paper, and then touch on some of the comments and criticisms that have been raised about it in various blogs.
Optimists, pessimists, and agnostics: Revisiting the three housing-boom camps
The paper basically asks a simple question: What did professional economists think about the housing boom while it was in full swing? Did most of them warn that the boom was really a bubble, sustained only by its own momentum and destined to crash someday? Or did most professional economists think that the run-up in housing values was entirely justified by fundamental factors, like low interest rates or higher rents? Our paper mainly discusses, rather than evaluates, the housing literature at this time. As a result, the paper is really more of a literature review rather than formal economic research. But we think that this type of review is helpful if people think that a disruptive gyration in asset prices could happen again.
By looking at the research papers and opinion pieces of professional economists, we concluded that these economists could be separated into three camps. All three camps agreed that house prices had risen dramatically—that was obvious from the data—but they differed on why. The "optimist" camp argued that there were good reasons why house prices had gone up, while the "pessimists" argued that there were no good reasons. A third group, whom we label "agnostics," refused to take a position one way or the other. Remarkably, most economists fell into this third camp. We argue that one reason for the popularity of the agnostic position is that economists generally believe that asset prices reflect fundamental factors unless there is a lot of strong evidence to the contrary. After all, asset prices are set in free markets, by people using their own money. Who are economists to say that the prices that people have agreed upon are wrong?
The strongest forms of this view comprise the various flavors of efficient-market theory, which argues that asset prices are always correct in some sense. While most economists would not go that far, they would still demand a lot of evidence before they would label some asset-price increase a bubble. This tendency to view asset prices as reflecting all available information makes economists generally reluctant to predict where asset prices are headed, or to say that some asset price is unsustainable. As we write in the paper:
In a sense, this reluctance to commit should not surprise anyone familiar with modern asset-pricing theory. The "Fundamental Theorem of Asset Pricing" implies that the evolution of asset prices is, to a first approximation, unpredictable. If housing was so obviously overvalued, as the pessimists suggested, then investors stood to make huge profits by betting against housing. By doing so, investors would have ensured that house prices would have fallen immediately. Regardless of whether the theory of the unpredictability of asset prices is correct, the theory is part of the basic training of almost every economist. Consequently, any economist who suggests to his or her peers that an asset is over- or undervalued faces a heavy burden of proof.
The optimists were reasonable people
Among bloggers, the paper has generated a mixed reaction. Naturally, we think the people that like the paper are absolutely correct in their assessments. But more seriously, we also think that our critics fail to appreciate the major point we were trying to make. In fact, some of their arguments actually provide further support for our position.
First off, many critics have said that housing-market optimists were right-wing ideologues who were deluded by efficient-market theory into thinking that asset bubbles were impossible. This story simply doesn't square with the facts. Two of the most prominent optimists were Chris Mayer and Todd Sinai. Both are professors at top business schools (Columbia and Wharton respectively), and both serve as co-heads of the Real Estate program at the National Bureau of Economic Research, a research body that includes economists of all ideological stripes. Even more importantly, one of Chris Mayer's seminal real-estate papers discusses "loss aversion" among potential house sellers. Loss aversion is a behavioral theory that blends psychology and economics, so it is not exactly high on the right-wing ideological research agenda, which is firmly rooted in the free market, rational decision-making tradition. By virtually any definition, Chris Mayer and Todd Sinai are reasonable people. Even so, based on their joint research, Mayer identified people who were predicting a collapse in house prices at the end of 2005 as "Chicken Littles."
The second issue has to do with economics as a science. Paul Krugman argues that "the evidence just screamed bubble." He points to the following picture, which shows real housing prices from the late 1970s through 2005.
This picture might convince the lay person that housing prices were headed for a fall. But the point of our paper is that convincing economists of this fact would have required much more work. Despite what some influential commentators might have us believe, large and persistent increases in real house prices have occurred in the U.S. without subsequent crashes. For example, Robert Shiller has found that U.S. housing prices increased by 60 percent between 1942 and 1947, adjusted for inflation. (This increase basically reversed a long-lasting period of low housing prices that occurred during the Depression.) After 1947, house prices were basically flat until the close of the 20th century. Nobody refers to the 1940s period as a house price bubble, because housing prices remained high for a very long time.
Another example of a long-lasting swing in housing prices is the boom–bust cycle along the U.S. coasts during the late 1980s and early 1990s. Krugman's post states that the late-'80s price rise in southern California depicted in this figure was a bubble, but that claim is debatable. Even after their recent crash, California housing prices are approximately 20 percent higher than they were during their late-1980s peak.
The responsibility of policymakers to create a robust financial system
What is most puzzling to us is the statement by some bloggers that we are essentially trying to exonerate the economics profession for failing to miss the housing bubble. Exactly the opposite is true. What we try to argue in the paper is that the economics profession simply doesn't have the tools to achieve real-time consensus on whether a bubble exists in this or that asset market. One response to this problem is for economists to relax their views on the correctness of asset prices. Undoubtedly, the recent housing cycle will nudge a lot of economists away from efficient-markets theory and the fundamental unpredictability of asset prices. For policymakers, we would argue the main lesson of our literature review is that they should not depend on economists to achieve consensus on whether some asset price reflects a potentially disruptive bubble. Rather, policymakers should construct a financial system that is robust enough to withstand steep drops in asset prices. If our review of the housing literature is any indication, there will be little warning of the next asset-market crash in the peer-reviewed economics literature.
By Kristopher Gerardi, research economist and assistant policy adviser at the Atlanta Fed, and Christopher Foote and Paul Willen, research economists and policy advisers at the Boston Fed