Housing bubbles can be devastating. The 2008 housing crisis provided a painful lesson about what can happen when house prices spiral out of control. When home prices rise far beyond what people can afford only to crash later, the fallout hurts families, communities, businesses and entire economies. In contrast to financial markets like the stock market, this a market to which everybody is exposed to some extent. But what if we could spot a bubble before it bursts? Detecting housing bubbles as they are happening has always been difficult. There is a lot of uncertainty surrounding what the price of housing should be. The broad aim of the International Housing Observatory is to increase our understanding of how housing markets operate. We have a specific focus on monitoring housing markets for periods of exuberance, of overheating. We are also trying to provide information to the public and to policymakers in real time. Economists traditionally look at fundamentals – factors such as income, population growth, construction costs, and interest rates – to determine what homes should reasonably cost. When prices climb far above what these factors suggest, it might signal a bubble. But nobody can agree on exactly which fundamentals matter most or how to measure them. One economist might focus on rental prices, another on local job growth, a third on mortgage rates. Models also rely on backward-looking data, are slow to update, and often miss the bigger picture. This makes it nearly impossible to definitively say whether rising prices reflect genuine economic conditions or dangerous speculation. A NEW WAY We took a different approach. In our research, we are trying to see what consumers believe about the housing market, to make use of their expectations about house prices. Using data from the University of Michigan Survey of Consumers (UMSC), we looked at what ordinary Americans expect to happen to housing prices. The UMSC has been asking Americans about their expectations for housing prices for decades. By comparing these forecasts to what happened, we can measure how accurate expectations were, and whether they consistently underestimated price growth. By looking at deviations between what people expect and what actually happens we get a clean signal of what is happening in the market, and whether we are in a bubble period or not. We are not looking at how big the deviation between belief and actuality is on its own; we are looking at how fast this deviation changes over time. If people keep under-guessing how fast home prices will rise – and that mistake snowballs month after month – it is a tell-tale sign that prices are being driven by hype rather than fundamentals. So, why should we look at “forecast errors”? Imagine a weather app that keeps saying, “Expect 20°C tomorrow,” yet the thermometer turns out to be 21°C, then 23°C, then 26°C. Each day is not just warmer than forecasted, but hotter than the day before. Those escalating misses pile up into a clear signal that something abnormal is driving temperatures beyond expectations. One of the advantages of our method compared to the traditional methods is that we have higher frequency data for survey expectations compared to those traditional fundamentals, like income, that we have every quarter. For real-time monitoring, this is a beneficial method to see what people expect about the housing market and how far away that is from what actually happens. TRACKING EXPECTATION This is how it works. Imagine that in 2019, survey respondents expected home prices to rise by 3% over the next year, but prices actually rose by 8%. Now imagine this pattern repeats month after month, with people consistently underestimating how fast prices are rising. This suggests prices might be climbing faster than even optimistic buyers think is reasonable, a potential sign of speculative excess. Using this method, we identified two major periods of housing market exuberance in the USA since the 2008 financial crisis. 20 |
RkJQdWJsaXNoZXIy NTI5NzM=