Given that the Congressional Budget Office kindly provides an historical series of both potential GDP and the NAIRU (non-accelerating inflation rate of unemployment), aka the lowest unemployment rate they judge to be consistent with stable inflation, we can compare those to the actual outcomes and measure slack over the ages.
That is, define every quarter that actual GDP is lower than potential GDP and every quarter that actual unemployment is above the NAIRU as a slack quarter—a three-month period when the overall economy and the job market were operating below full strength. The underlying models for these indicators are similar but not the same. For example, going into a recession, GDP might stumble before the jobless rate begins to rise, leading to a quarter when GDP is below potential but unemployment is still below the NAIRU. But most of the time, the two indicators will agree.
The first two figures plot 1) the ratio of actual GDP to potential GDP and 2) the unemployment rate minus the NAIRU. When the former is more than 1, real GDP is above potential; when the latter is below zero, the job market is tight.
Sources: CBO, BLS, BEA
Eyeballing the charts, 1979 might be viewed as a dividing line between operating without slack most of the time and vice versa. This next figure thus asks “what percent of the time was GDP too low or unemployment too high?” The answers, as you see below, point to much more slack in the latter period. GDP was below potential 42% of the quarters in the earlier periods compared to 74% in the latter; the comparable percentages for too-high-unemployment are 29% and 69%, a huge difference in the persistence of slack labor markets.
Sources: Same as above.
I will also note that it is post-1979 that we’ve seen much more inequality, wage stagnation, sticky poverty rates, and a generalized disconnect between overall growth and the living standards of many middle class and poor households.
Explaining these differences is beyond the scope of one blog post. Globalization is oft-cited, and it is the case that net exports as a share of GDP averaged about zero in the first period and about -2.5% in the latter. And no question that austere fiscal policy has played an important role in dampening both GDP and job growth in recent years. There are also measurement issues: potential GDP and the NAIRU are estimates and the analysis is sensitive to how they’re derived. But I strongly suspect any such analytic exercise would yield similar results.
So what’s the point of all this? Good question. For me, at least, it’s not: let’s go back to the 1950s! Instead, it’s let’s be very much aware that the assumption that we’re typically at full employment and the economy’s hitting on all cylinders is indefensible, or, at the very least, inconsistent with the empirical record.
To be a bit more specific, as an example of how these pictures might be used, consider the argument that the problem is some people just don’t want to work, and all it will take for them to get a job is to cut their UI or SNAP benefits. Based on what you see here, that’s an argument based on wishful thinking at best and engrained prejudice at worst. But it’s not based on data.
By these metrics, a model of the economy that weighted market failures above market auto-efficiencies would have a far better track record. If we were physicians and these figures were in the patient’s chart at the end of her hospital bed, we’d call for a consult of the medical team and a serious revaluation of the course of treatment.