August 15, 2016
Robert Samuelson used his column this week to note my friend Josh Biven’s piece on the reason for the weak recovery. Biven puts the blame on insufficient government spending, noting that government spending per capita has been much weaker in this recovery than in prior recoveries. Samuelson says Biven could be right but then argues that maybe there is another explanation.
Samuelson offers the possibility that higher government spending in past downturns may have been the result of more rapid economic growth rather than the cause. He notes that the stimulus failed to lead to sustained growth in this recovery, despite its large size. He then offers three possible explanations:
“Some economists see a broad slowdown in technological advances (despite the Internet) whose adverse effects were masked by easy credit. Another theory is that the costs of the welfare state and regulation have come home to roost; they allegedly discourage risk-taking, business investment and work. Another view is that the financial crisis and the Great Recession so scared consumers and businesses that they are reluctant to spend.”
Let’s deal with these issues in turn. First, Samuelson does have a reasonable point on the cause and effect story. State and local governments were seeing more rapid revenue growth in prior recoveries, so it would not be surprising that their spending grew more rapidly. Bivens is undoubtedly right that austerity at all levels of government slowed growth, but the issue is not quite as simple as it first appears.
On Samuelson’s three points, he is right that many economists point to a broad slowdown in technological advances. While this is a very big issue, one point that should bother anyone taking it seriously is that the slowdown seems to have hit most of the world at the same time. Some countries, like the United States, were much further along in adopting the new technologies of the prior decade than countries like Greece. The fact that we all are experiencing a productivity slowdown at the same time nonetheless suggests that it is not the lack of technology that is the problem.
The welfare state story doesn’t fit well for the same reason. The problems of the welfare states suddenly hit the U.S. in 2010, but not 10 or 20 years earlier. What happened in 2010 and why didn’t the problem hit other countries with much more generous welfare states much earlier. Okay, Samuelson doesn’t like the welfare state, but this story doesn’t make any sense.
Finally, we get the spectre of the financial crisis haunting consumers and businesses. The problem with this one is that consumers are now spending a very high share of their income — no evidence of haunting there. And businesses are spending almost as much on investment, measured as a share of GDP, as they did before the downturn. So the evidence doesn’t fit the story.
So what else can explain the weak downturn? How about something incredibly simple, like we had a big housing bubble driving the economy and there is nothing to replace the demand created by the bubble?
The bubble pushed residential construction to more than 6.0 percent of GDP at its peak in 2005. It is still under 4.0 percent of GDP. That means it is down by around 2.5 percentage points of GDP ($450 billion a year in today’s economy). In addition, the housing wealth effect led to an enormous consumption boom. When the wealth disappeared, the housing-wealth-based consumption disappeared also — surprise, surprise, surprise. If consumption were the same share of income as at the peak of the bubble, it would be about 2.0 percentage points of GDP higher ($360 billion a year).
So, the story is that we have nothing to replace $450 billion a year in lost construction demand and $360 billion a year in lost consumption demand. That is a total of $810 billion a year in lost demand — that is before counting any multiplier effects. So, why is it a surprise that we haven’t seen a robust recovery?
All past recessions, with the exception of the 2001 recession that was caused by the collapse of the stock bubble, were the result of the Fed raising interest rates to combat inflation. Raising interest rates reduces demand for houses and cars. The good part of this story is that the temporary reduction in demand creates a large amount of pent-up demand. This means that when the Fed lowers interest rates it leads to an explosion in residential construction and car purchases. This gets the economy back in gear and leads to a sustained growth path.
This sort of explosion in demand couldn’t happen this time because the recession wasn’t caused by the Fed raising interest rates, it was caused by a collapse of a bubble. Instead of pent-up demand, there was an enormous excess supply of housing. We simply didn’t have the basis for a normal recovery.
I know this is far too simple for Robert Samuelson. In fact, it is far too simple for most economists who couldn’t see the $8 trillion housing bubble that crashed the economy because it was far too simple. But this is the story that makes sense, unlike the ones he is pushing in his piece.
One other point that is worth mentioning. We could fill the gap in demand by reducing the trade deficit, which is now a bit less than 3.0 percent of GDP (@ $500 billion a year). This is again intro econ type stuff, but the folks at the Washington Post call you names if you bring accounting identities into discussions of economic policy.
Addendum:
Just to throw in some numbers here, residential construction rose by 25.2 percent from the first quarter of 1975 to the first quarter of 1976. Durable goods consumption grew by 16.9 percent. Residential construction grew by 31.5 percent from the first quarter of 1983 to the first quarter of 1984. Durable goods consumption rose by 21.4 percent. The strongest period of housing growth was 12.6 percent from the first quarter of 2012 to the first quarter of 2013. Durable good consumption grew by just 6.9 percent over this period.
Note also that this period of rapid housing growth began almost three years after the start of the recovery. In the other recoveries, the spurt of residential construction occurred at the beginning of the recovery.
Comments