FacebookTwitterRedditEmail

The NYT Prints Its Usual Nonsense About Bubbles

The NYT has apparently assigned Ruchir Sharma the task of writing periodic pieces about the prospect of a crashing bubble giving us another horrible recession. These pieces always show a failure to understand the most basic features of the Great Recession. This fits with the need of elite-types to pretend that the risks of the housing bubble were hard to see, as opposed to requiring a passing glance at quarterly GDP data.

Sharma’s latest tells us that we should be worried because house prices are rising even as we are in a recession. While it is true that house prices are rising, people who have been paying attention to the data, know that large segments of the population are doing just fine, in spite of the recession. The job loss has been hugely concentrated among those in relatively low-paying industries, like hotels and restaurants. These lower-paid workers are much less likely to be home buyers than the workers who kept their jobs.

With interest rates at historic lows, people can afford to pay more for housing, as Sharma notes. And, with many more workers now able to work remotely, it should not be surprising that we would see a strong housing market.

Sharma implies that we would face some catastrophic situation if interest rates and then the prices of houses and other assets fall. People who have access to the Commerce Department’s GDP data know that Sharma doesn’t have much of a case. In the housing bubble before the Great Recession, housing construction peaked at 6.7 percent of GDP in 2005. After the collapse of the bubble, it bottomed out at less than 2.0 percent of GDP. This implied a loss in demand of 4.7 percentage points of GDP, which would be equivalent to roughly $1 trillion in annual demand in today’s economy.

In addition, there was also a plunge in consumption of more than 2.0 percent of GDP following the crash. The bubble had led to a record consumption boom, as the savings rate fell to less than 4.0 percent of disposable income. This bubble driven consumption disappeared when the bubble burst and the savings rate returned to more normal levels.

The fact that we would see a serious recession following the collapse of the housing bubble was 100 percent predictable for anyone who follows the basic GDP data. There is no remotely comparable story today. Housing was 4.2 percent of GDP in the third quarter, only slightly higher than its long-term average. Furthermore, unlike during the bubble years, we are not seeing extraordinarily high vacancy rates in most areas.

Also, in contrast to the bubble years, consumption is relatively low, as the savings rate is near record highs. A plunge in house prices may upset some homeowners but would likely have little impact on consumption. In short, there is no basis for the concerns in Sharma’s column.

It’s worth adding that the collapse of the stock bubble in the years 2000-2002 was able to cause a recession because it too was driving the economy. Investment, especially in the tech sector, rose sharply at the end of the 1990s, as companies with no profits, and no idea how they could make a profit, were able to raise billions by issuing stock. The stock wealth generated by the bubble also led to a consumption boom. These were reversed when the bubble collapsed.

While the 2001 recession is usually considered short and mild, this is not true from the perspective of the labor market. It took the economy four full years to recover the jobs lost in the downturn. At the time, this was the longest period without job growth since the Great Depression.

This first appeared in Dean Baker’s Beat the Press blog.

Dean Baker is the senior economist at the Center for Economic and Policy Research in Washington, DC. 

FacebookTwitterRedditEmail