34 United States The EconomistFebruary 10th 2018
12 cy since the financial crisis would create
anotherbubble—one that, they might say,
has nowbegun to deflate. Butunlike in the
1990s, no single major asset class is inspir-
ing irrational exuberance. From bonds to
buildings, a wide range ofassets are expen-
sive. High prices can be justified so long as
low interest rates make future flows of in-
come look more valuable.
In theory, fiscal stimulus should force
interest rates up, which explains investors’
worries. The Fed’s model predicts that a
tax cut worth 1% ofGDPwill eventually
raise rates by 0.4 percentage points. If the
central bank tries to keep rates where they
were before the stimulus, the theory goes,
inflation will get out of hand. But there are
three reasons to doubt that, in reality,
enough inflation will appear to force the
Fed to change course.
The first is the prospect of another pro-
ductivity surge. Productivity growth has
been feeble everywhere since the financial
crisis. Yet technological evangelists insist
that a second industrial revolution is com-
ing, in which machine learning and artifi-
cial intelligence will allow firms to do
much more with fewer workers. Whether
they are right or not, rising wages should
encourage firms to invest more in labour-
saving technology. There are signs already
that productivity is rebounding. In every
quarter of 2017 it was more than 1% higher
than a year before, the first such sustained
growth since 2010.
The trend seems to be continuing this
year. A real-time estimate of annualised
GDP growth in the first quarter of 2018 by
the Atlanta Fed stands at 4%. If this esti-
mate is even close to correct, it points to
strong productivity growth. The alterna-
tive explanation is an unusual rise in em-
ployment or average hours worked. But
job growth has in fact slowed in recent
years; in January it was barely above the
average for 2016, when the economy grew
by just 1.8%. Average hours worked have
fallen slightly.
The second reason to expect inflation to
remain subdued is the painful legacy of
the financial crisis. Economistshave long
speculated that recessions might damage
the supply capacity of the economy. When
people are thrown out of work for months
or years, for example, their skills start to at-
rophy. If so, the thinking goes, strong
growth might arrest and even reverse this
process—perhaps by encouraging labour-
starved firms to offer more training to new
staff. That should raise productivity, limit-
ing inflationary pressure. In 2016 Janet Yel-
len, then chair of the Fed, wondered if the
possibility of expanding supply might jus-
tify running a “high-pressure economy”
once the recovery was complete.
Evidence that recessions damage work-
ers’ skills is patchy. But a related phenome-
non may be at work. Because the unem-
ployment rate excludes people who arenot seeking jobs, it could be masking po-
tential labour supply. In April 2000 nearly
82% of Americans aged between 25 and 54
had jobs. Today, despite low unemploy-
ment, the proportion is 79%. The difference
represents about 3.7m potential workers. It
is the result of a decline in labour-force par-
ticipation: many people are neither work-
ing nor looking for work.Get back to work
Official forecasters tend to assume that
participation trends are immutable, and
do not respond much to economic condi-
tions. But evidence to the contrary is build-
ing. In a recent working paper, Danny Ya-
gan of the University of California,
Berkeley, compares places where unem-
ployment rose a lot during the recession,
such as Phoenix, in Arizona, to those
where the increase was less severe, such as
San Antonio, in Texas. He finds that for ev-
ery one percentage-point rise in local un-
employment during the recession, work-
ing-age people were 0.4 percentage points
less likely to be employed in 2015. Unem-
ployment rates in these places have largely
converged again, whereas overall employ-
ment rates have not, suggesting that some
workers were so deterred that they left the
labour force altogether. Doves argue that
the lower unemployment falls, the greater
the chance of bringing such workers backinto the fold. Sure enough, since late 2015,
as the labour market has tightened, partici-
pation among prime-age workers has risen
sharply.
The final reason not to fear an inflation-
ary surge is the stability of wage and price
growth in recent years. Although markets
were spooked by a rise in hourly earnings,
perhaps they should not have been. Pay is
growing almost exactly as quickly as one
would expect from looking at the overall
employment rate (see chart 2). Neither
wages nor prices are likely to accelerate
suddenly, as some economic models fore-
cast, because low inflation expectations
have become so firmly rooted. In a recent
paper for the Peterson Institute, a think-
tank, Olivier Blanchard, who was until
2015 the IMF’s chief economist, concludes
that the current relationship between un-
employment and inflation is “at odds with
the accelerationist hypothesis”.
Higher wages need not mean higher
prices. In fact, economistshave recently
struggled to establish a causal link. Core in-
flation, excluding volatile food and energy
prices, stands at 1.5%, just over half the rate
of wage growth. Apart from higher produc-
tivity, one way the economy might absorb
higher pay is if firms’ profit margins shrink.
Over recent decades, corporate profits
have risen to record highs as a percentage
ofGDP. Meanwhile the share of national2 3Expect the expectedSources: Moody’s Analytics; Bureau of Labour StatisticsUnited States*1994 onwards †25- to 54-year-olds01234562010 11 12 13 14 15 16 1725th
percentileMedian75th
percentileEmployment as % of population†Private wages and salaries,% increase on a year earlier
01234574 75 76 77 78 79 80 81 821990s* 2000s 2010s 2017Q4 2017Weekly earnings, full-time employees
By income percentile, % increase on a year earlierWages and employment, quarterly